{"object":"list","data":[{"id":"openai/gpt-6-astra","providers":["azure-foundry","openai"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":58.5,"artificial_analysis_coding_index":76.9,"artificial_analysis_agentic_index":51,"artificial_analysis_economics_index":59.8,"artificial_analysis_engineering_index":54.7,"artificial_analysis_intelligence_index":52.7,"artificial_analysis_strategy_and_ops_index":57.3,"artificial_analysis_finance_and_accounting_index":55.2,"artificial_analysis_healthcare_and_medical_index":51.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"modelCount":5,"intelligenceIndex":52.673669395513},"name":"GPT-6 Astra","slug":"gpt-6-astra","source":"https://artificialanalysis.ai/models/gpt-6-astra","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-6-astra/benchmarks"},"created":0,"released_at":"2026-09-03","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-6 Astra is OpenAI's flagship model, built for the hardest end-to-end work across complex reasoning, coding, computer use, research, and document creation. It accepts text and image input, offers configurable reasoning effort up to a max setting, and pairs with web search, file search, code execution, and shell tools.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-6-astra","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-6-astra","title":"OpenAI model documentation"}],"display_name":"GPT-6 Astra","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":10,"output_per_mtok_usd":50,"cached_input_per_mtok_usd":1,"cache_write_per_mtok_usd":12.5,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":20,"output_per_mtok_usd":75,"cached_input_per_mtok_usd":2,"cache_write_per_mtok_usd":25}]}},{"id":"openai/gpt-6.1-sol","providers":["azure-foundry","openai"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":57.7,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":59.3,"artificial_analysis_engineering_index":53.9,"artificial_analysis_intelligence_index":51.8,"artificial_analysis_strategy_and_ops_index":57,"artificial_analysis_finance_and_accounting_index":54,"artificial_analysis_healthcare_and_medical_index":51.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"modelCount":5,"intelligenceIndex":51.8332597011541},"name":"GPT-6.1 Sol","slug":"gpt-6-1-sol","source":"https://artificialanalysis.ai/models/gpt-6-1-sol","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-6.1-sol/benchmarks"},"created":0,"released_at":"2026-09-29","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-6.1 Sol is OpenAI's near-flagship model, released a week after GPT-6 Sol and aimed at agentic coding, computer use and other demanding professional work. OpenAI positions it as delivering close to GPT-6 Astra's quality at a fraction of Astra's token prices, and it reasons on a narrower effort ladder than its predecessor.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"},{"url":"https://techcrunch.com/2026/09/29/openai-launches-gpt-6-1-sol-says-it-nearly-matches-gpt-6-astra-and-costs-less/","title":"TechCrunch: OpenAI launches GPT-6.1 Sol"}],"display_name":"GPT-6.1 Sol","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.1,"cache_write_per_mtok_usd":2.5,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":4,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.2,"cache_write_per_mtok_usd":5}]}},{"id":"openai/gpt-6.1-terra","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":49,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":53.6,"artificial_analysis_engineering_index":49.1,"artificial_analysis_intelligence_index":48.1,"artificial_analysis_strategy_and_ops_index":49.5,"artificial_analysis_finance_and_accounting_index":47.1,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-26T19:50:56.616Z"},"data":{"modelCount":6,"intelligenceIndex":48.1061836463044},"name":"GPT-6.1 Terra","slug":"gpt-6-1-terra","source":"https://artificialanalysis.ai/models/gpt-6-1-terra","fetchedAt":"2026-10-26T19:50:56.616Z","pageRecords":[]}],"checkedAt":"2026-10-26T19:50:56.616Z","fetchedAt":"2026-10-26T19:50:56.616Z","nextCheckAt":"2026-11-02T19:50:56.616Z","summary":true,"detailPath":"/v1/models/openai/gpt-6.1-terra/benchmarks"},"created":0,"released_at":"2026-10-26","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-6.1 Terra is OpenAI's fast, efficient mid-tier model in the GPT-6.1 family, sitting beneath GPT-6.1 Sol. It targets visual reasoning, precise computer use and clear, natural writing, and revives the Terra tier that balanced capability and cost between Sol and Luna in the GPT-5.6 generation.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-6.1-terra","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"},{"url":"https://www.testingcatalog.com/openai-prepares-gpt-6-1-terra-microsoft-code/","title":"TestingCatalog: OpenAI prepares GPT-6.1 Terra for the upcoming release"}],"display_name":"GPT-6.1 Terra","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.05,"cache_write_per_mtok_usd":1.25,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":2,"output_per_mtok_usd":7.5,"cached_input_per_mtok_usd":0.1,"cache_write_per_mtok_usd":2.5}]}},{"id":"openai/gpt-6-sol","providers":["azure-foundry","openai"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":51.1,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":54.4,"artificial_analysis_engineering_index":48.6,"artificial_analysis_intelligence_index":47.6,"artificial_analysis_strategy_and_ops_index":52.5,"artificial_analysis_finance_and_accounting_index":48.6,"artificial_analysis_healthcare_and_medical_index":43.6}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.479147358665431,"lcr":0.836666666666667,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.308571428571429,"gdpval":1508.23,"aa3dElo":1123.3587,"ifbench":null,"mmmuPro":0.829479768786127,"scicode":0.576388888888889,"harveyLab":0.03611111111111111,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.494350282485876,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"mlcrOverall":0.338888888888889,"omniscience":27.116666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"gdpPdfAllPass":0.252,"livecodebench":null,"hostModelCount":7,"cacheWritePrice":2.5,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":0.439393939393939,"gdpvalNormalized":0.504115,"intelligenceIndex":47.6305176437724,"terminalbenchHard":null,"briefcaseTotalCost":238.8123648528785,"price1mInputTokens":2,"contextWindowTokens":1050000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.5448333333333333,"price1mOutputTokens":10,"pricePer1k1mpImages":null,"terminalBenchScience":0.3,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":6,"price1mBlended0To3To1":4,"price1mBlended7To2To1":1.54,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.0792079207920793,"price1mBlended100To1To1":0.3137254901960784,"automationBenchPartialScore":0.6163369295959864,"intelligenceIndexTimePerTask":289.39748102176264,"omniscienceHallucinationRate":0.6012449652142072,"inferenceParametersActiveBillions":null},"name":"GPT-6 Sol (Max)","slug":"gpt-6-sol","source":"https://artificialanalysis.ai/models/gpt-6-sol","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]},{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":null,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":null,"lcr":null,"gpqa":null,"tau2":null,"aime25":null,"critpt":null,"gdpval":null,"ifbench":null,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":2.5,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":null,"terminalbenchHard":null,"price1mInputTokens":2,"contextWindowTokens":1050000,"knowledgeCutoffDate":null,"price1mOutputTokens":10,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":6,"price1mBlended0To3To1":4,"price1mBlended7To2To1":1.54,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.0792079207920793,"price1mBlended100To1To1":0.3137254901960784,"inferenceParametersActiveBillions":null},"name":"GPT-6 Sol (Daybreak Blue, max)","slug":"gpt-6-sol-daybreak-blue","source":"https://artificialanalysis.ai/models/gpt-6-sol-daybreak-blue","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-6-sol/benchmarks"},"created":0,"released_at":"2026-09-22","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-6 Sol is the cost-efficient high-end tier of OpenAI's GPT-6 family, sitting below the flagship GPT-6 Astra and above the smaller GPT-6 Luna. It is aimed at demanding professional work — agentic coding in real repositories, computer use and long-horizon workflow automation — where Astra-class reliability is wanted without Astra's price.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-6-sol","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"},{"url":"https://openrouter.ai/openai/gpt-6-sol","title":"OpenRouter model listing"}],"display_name":"GPT-6 Sol","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.2,"cache_write_per_mtok_usd":2.5,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":4,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.4,"cache_write_per_mtok_usd":5}]}},{"id":"openai/gpt-6-luna","providers":["azure-foundry","openai"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":40.3,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":45.2,"artificial_analysis_engineering_index":37.1,"artificial_analysis_intelligence_index":38.1,"artificial_analysis_strategy_and_ops_index":45.7,"artificial_analysis_finance_and_accounting_index":40.3,"artificial_analysis_healthcare_and_medical_index":37.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"modelCount":6,"intelligenceIndex":38.1245186869738},"name":"GPT-6 Luna","slug":"gpt-6-luna","source":"https://artificialanalysis.ai/models/gpt-6-luna","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-6-luna/benchmarks"},"created":0,"released_at":"2026-09-22","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-6 Luna is the small, cost-efficient tier of OpenAI's GPT-6 family, sitting below GPT-6 Sol and GPT-6 Astra. It targets high-volume, latency-sensitive work such as chat, classification and lightweight agents, and at higher reasoning effort it takes on tasks that previously needed a Sol-tier model.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-6-luna","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"},{"url":"https://openrouter.ai/openai/gpt-6-luna","title":"OpenRouter model listing"}],"display_name":"GPT-6 Luna","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0.5,"cached_input_per_mtok_usd":0.01,"cache_write_per_mtok_usd":0.125,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":0.2,"output_per_mtok_usd":0.75,"cached_input_per_mtok_usd":0.02,"cache_write_per_mtok_usd":0.25}]}},{"id":"openai/gpt-6-luna-decisions","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T18:34:54.689Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/openai/gpt-6-luna-decisions/benchmarks"},"created":0,"released_at":"2026-10-06","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-6 Luna Decisions is GPT-6 Luna served through OpenAI's Decisions API: it reads text or images against typed yes/no, choice, and score questions and returns probabilities and confidence instead of generated text. It is built for classification, routing, and choosing an agent's next action in a single fast call, and bills input tokens only.","description_sources":[{"url":"https://developers.openai.com/api/docs/guides/decisions","title":"OpenAI Decisions guide"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"}],"display_name":"GPT-6 Luna Decisions","context_window":1050000,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/decisions"],"input_modalities":["text","image"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":0.2,"output_per_mtok_usd":0}]}},{"id":"openai/gpt-5.6-sol","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":49.8,"artificial_analysis_coding_index":77.4,"artificial_analysis_agentic_index":50.2,"artificial_analysis_economics_index":52.9,"artificial_analysis_engineering_index":48.9,"artificial_analysis_intelligence_index":47,"artificial_analysis_strategy_and_ops_index":54.3,"artificial_analysis_finance_and_accounting_index":49,"artificial_analysis_healthcare_and_medical_index":45.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.494902687673772,"lcr":0.84,"gpqa":0.941414141414141,"tau2":0.850877192982456,"aime25":null,"critpt":0.322857142857143,"gdpval":1608.63,"ifbench":0.726530612244898,"mmmuPro":0.834104046242775,"scicode":0.570601851851852,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.562146892655367,"licenseUrl":null,"parameters":null,"tauBanking":0.443298969072165,"licenseName":null,"mlcrOverall":0.3,"omniscience":21.966666666666665,"analystAgent":0.475,"frontierCode":null,"cacheHitPrice":0.4,"gdpPdfAllPass":0.272,"livecodebench":null,"hostModelCount":8,"cacheWritePrice":5,"reasoningTokens":0,"terminalBench21":0.880149812734082,"terminalBench40":0.398989898989899,"enterpriseOpsGym":0.42912563413906296,"gdpvalNormalized":0.554315,"intelligenceIndex":46.9727068915036,"terminalbenchHard":0.659090909090909,"briefcaseTotalCost":365.8439549352594,"price1mInputTokens":4,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.594,"price1mOutputTokens":20,"pricePer1k1mpImages":null,"terminalBenchScience":0.223809523809524,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":12,"price1mBlended0To3To1":8,"price1mBlended7To2To1":3.08,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":4.158415841584159,"price1mBlended100To1To1":0.6274509803921569,"automationBenchPartialScore":0.6008114996276329,"intelligenceIndexTimePerTask":347.5884115016372,"omniscienceHallucinationRate":0.9220032840722496,"inferenceParametersActiveBillions":null},"name":"GPT-5.6 Sol (Max)","slug":"gpt-5-6-sol","source":"https://artificialanalysis.ai/models/gpt-5-6-sol","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.6-sol/benchmarks"},"created":0,"released_at":"2026-07-09","release_source":"https://openai.com/index/gpt-5-6/","description":"GPT-5.6 Sol is the flagship tier of OpenAI's GPT-5.6 series, positioned for complex professional work such as multi-step reasoning, coding, and agentic workflows. It corresponds to the unsuffixed tier of earlier GPT-5 generations, with text and image input, configurable reasoning effort, and computer-use and code-interpreter tools.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.6-sol","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol","title":"OpenAI model documentation"}],"display_name":"GPT-5.6 Sol","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":4,"output_per_mtok_usd":20,"cached_input_per_mtok_usd":0.4,"cache_write_per_mtok_usd":5,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":8,"output_per_mtok_usd":30,"cached_input_per_mtok_usd":0.8,"cache_write_per_mtok_usd":10}]}},{"id":"openai/gpt-5.6-luna","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":36.9,"artificial_analysis_coding_index":71.4,"artificial_analysis_agentic_index":42.1,"artificial_analysis_economics_index":42.7,"artificial_analysis_engineering_index":36.2,"artificial_analysis_intelligence_index":37.3,"artificial_analysis_strategy_and_ops_index":44.1,"artificial_analysis_finance_and_accounting_index":40.1,"artificial_analysis_healthcare_and_medical_index":35.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.394810009267841,"lcr":0.836666666666667,"gpqa":0.911111111111111,"tau2":null,"aime25":null,"critpt":0.205714285714286,"gdpval":1459.2,"ifbench":null,"mmmuPro":0.785549132947977,"scicode":0.53587962962963,"apexAgents":0.358407079646018,"hleDiamond":null,"itBenchSre":0.403201506591337,"licenseUrl":null,"parameters":null,"tauBanking":0.311340206185567,"licenseName":null,"mlcrOverall":0.2,"omniscience":-10.283333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.02,"gdpPdfAllPass":0.24,"livecodebench":null,"hostModelCount":7,"cacheWritePrice":0.25,"reasoningTokens":0,"terminalBench21":0.808988764044944,"terminalBench40":0.116161616161616,"enterpriseOpsGym":0.40823634735899733,"gdpvalNormalized":0.4796,"intelligenceIndex":37.3244239690841,"terminalbenchHard":null,"briefcaseTotalCost":35.950831709230854,"price1mInputTokens":0.2,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.42733333333333334,"price1mOutputTokens":1.2,"pricePer1k1mpImages":null,"terminalBenchScience":0.0333333333333333,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.7,"price1mBlended0To3To1":0.45,"price1mBlended7To2To1":0.174,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.2099009900990099,"price1mBlended100To1To1":0.03333333333333334,"automationBenchPartialScore":0.5020861763756194,"intelligenceIndexTimePerTask":309.71607397433024,"omniscienceHallucinationRate":0.9257857974388825,"inferenceParametersActiveBillions":null},"name":"GPT-5.6 Luna (Max)","slug":"gpt-5-6-luna","source":"https://artificialanalysis.ai/models/gpt-5-6-luna","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.6-luna/benchmarks"},"created":0,"released_at":"2026-07-09","release_source":"https://openai.com/index/gpt-5-6/","description":"GPT-5.6 Luna is the fast, cost-efficient tier of OpenAI's GPT-5.6 series, designed for high-volume chat, classification, and lightweight agentic tasks. It roughly corresponds to the nano tier of earlier GPT-5 families while keeping text and image input, adjustable reasoning effort, and tool use.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.6-luna","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna","title":"OpenAI model documentation"}],"display_name":"GPT-5.6 Luna","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.2,"output_per_mtok_usd":1.2,"cached_input_per_mtok_usd":0.02,"cache_write_per_mtok_usd":0.25,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":0.4,"output_per_mtok_usd":1.8,"cached_input_per_mtok_usd":0.04,"cache_write_per_mtok_usd":0.5}]}},{"id":"openai/gpt-5.6-terra","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":41.5,"artificial_analysis_coding_index":76.7,"artificial_analysis_agentic_index":43.2,"artificial_analysis_economics_index":44.9,"artificial_analysis_engineering_index":43.1,"artificial_analysis_intelligence_index":42.1,"artificial_analysis_strategy_and_ops_index":47.5,"artificial_analysis_finance_and_accounting_index":41.4,"artificial_analysis_healthcare_and_medical_index":40.3}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"modelCount":6,"intelligenceIndex":42.0829248239494},"name":"GPT-5.6 Terra","slug":"gpt-5-6-terra","source":"https://artificialanalysis.ai/models/gpt-5-6-terra","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.6-terra/benchmarks"},"created":0,"released_at":"2026-07-09","release_source":"https://openai.com/index/gpt-5-6/","description":"GPT-5.6 Terra is the balanced tier of OpenAI's GPT-5.6 series, designed for workloads that weigh intelligence against cost in everyday coding, reasoning, and agentic tasks. It sits between the flagship Sol and the budget Luna, roughly matching the mini tier of earlier GPT-5 families, with configurable reasoning effort.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.6-terra","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra","title":"OpenAI model documentation"}],"display_name":"GPT-5.6 Terra","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":12,"cached_input_per_mtok_usd":0.2,"cache_write_per_mtok_usd":2.5,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":4,"output_per_mtok_usd":18,"cached_input_per_mtok_usd":0.4,"cache_write_per_mtok_usd":5}]}},{"id":"openai/gpt-image-2.5-sunburst","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/text-to-image/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"elo":1222,"ci_95":10},"name":"GPT Image 2.5 Sunburst (max) (text-to-image)","slug":"text-to-image:gpt-image-2.5-sunburst-2026-09-08","source":"https://artificialanalysis.ai/api/v2/media/text-to-image/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]},{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/image-editing/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"elo":1199,"ci_95":11},"name":"GPT Image 2.5 Sunburst (max) (image-editing)","slug":"image-editing:gpt-image-2.5-sunburst-2026-09-08","source":"https://artificialanalysis.ai/api/v2/media/image-editing/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-image-2.5-sunburst/benchmarks"},"created":0,"released_at":"2026-09-08","release_source":"https://openai.com/index/introducing-chatgpt-images-2-5/","description":"GPT Image 2.5 Sunburst is OpenAI's most capable image generation and editing model, positioned for work where editing precision matters most. It takes text and image input, supports inpainting and quality settings from low to max, and sits above the faster Flare sibling in the family.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-image-2.5-sunburst","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/guides/image-generation","title":"OpenAI image generation guide"}],"display_name":"GPT Image 2.5 Sunburst","context_window":32000,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":30}},{"id":"openai/gpt-image-2.5-flare","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/text-to-image/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"elo":1207,"ci_95":10},"name":"GPT Image 2.5 Flare (max) (text-to-image)","slug":"text-to-image:gpt-image-2.5-flare-2026-09-08","source":"https://artificialanalysis.ai/api/v2/media/text-to-image/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]},{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/image-editing/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"elo":1181,"ci_95":11},"name":"GPT Image 2.5 Flare (max) (image-editing)","slug":"image-editing:gpt-image-2.5-flare-2026-09-08","source":"https://artificialanalysis.ai/api/v2/media/image-editing/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-image-2.5-flare/benchmarks"},"created":0,"released_at":"2026-09-08","release_source":"https://openai.com/index/introducing-chatgpt-images-2-5/","description":"GPT Image 2.5 Flare is OpenAI's fastest model for high-quality, everyday image generation and editing. It takes text and image input with inpainting and selectable quality levels, trading some of the Sunburst sibling's editing precision for speed.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-image-2.5-flare","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/guides/image-generation","title":"OpenAI image generation guide"}],"display_name":"GPT Image 2.5 Flare","context_window":32000,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":30}},{"id":"openai/gpt-image-2","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/text-to-image/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"elo":1186,"ci_95":11},"name":"GPT Image 2 (high) (text-to-image)","slug":"text-to-image:gpt-image-2","source":"https://artificialanalysis.ai/api/v2/media/text-to-image/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]},{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/image-editing/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"elo":1136,"ci_95":13},"name":"GPT Image 2 (high) (image-editing)","slug":"image-editing:gpt-image-2","source":"https://artificialanalysis.ai/api/v2/media/image-editing/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-image-2/benchmarks"},"created":0,"released_at":"2026-04-21","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT Image 2 is OpenAI's image generation model for fast, high-quality image creation and editing from text and image prompts. It supports inpainting, flexible output sizes, and high-fidelity edits through the images endpoints, and it is the base that the GPT Image 2.5 Sunburst and Flare tiers build on.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-image-2","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/guides/image-generation","title":"OpenAI image generation guide"}],"display_name":"GPT Image 2","context_window":32000,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":30}},{"id":"openai/gpt-transcribe","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"aa_wer_index":0.033},"name":"GPT Transcribe, OpenAI (speech-to-text)","slug":"speech-to-text:d15963b1-a8b7-483c-94c0-d59a6260d53a","source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-transcribe/benchmarks"},"created":0,"released_at":"2026-07-28","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT Transcribe is OpenAI's speech-to-text model for transcribing completed audio files, streamed file transcripts, and committed turns in Realtime sessions. It accepts unstructured context, keyword hints, and multiple language hints to improve accuracy on domain-specific terminology, multilingual audio, and code-switching.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-transcribe","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/guides/speech-to-text","title":"OpenAI speech-to-text guide"}],"display_name":"GPT Transcribe","output_modality":"audio_transcription","input_modalities":["audio","text"],"output_modalities":["text"],"endpoints":["/v1/audio/transcriptions"],"max_file_size_bytes":26214400,"pricing":{"audio_input_per_minute_usd":0.0045}},{"id":"openai/text-embedding-3-small","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T18:34:54.689Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/openai/text-embedding-3-small/benchmarks"},"created":0,"released_at":"2024-01-25","release_source":"https://openai.com/index/new-embedding-models-and-api-updates/","description":"Text Embedding 3 Small is OpenAI's compact embedding model, the successor to the ada embedding model for search, clustering, recommendations, and classification. It is the cost-efficient choice in the family, converting text into vectors for relatedness measurement below the large model's price tier.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/text-embedding-3-small","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/guides/embeddings","title":"OpenAI embeddings guide"}],"display_name":"Text Embedding 3 Small","context_window":8192,"reasoning":null,"output_modality":"embedding","output_modalities":["embedding"],"endpoints":["/v1/embeddings"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.02,"output_per_mtok_usd":0}},{"id":"openai/text-embedding-3-large","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T18:34:54.689Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/openai/text-embedding-3-large/benchmarks"},"created":0,"released_at":"2024-01-25","release_source":"https://openai.com/index/new-embedding-models-and-api-updates/","description":"Text Embedding 3 Large is OpenAI's most capable embedding model, converting text into vectors for search, clustering, recommendations, anomaly detection, and classification. It is positioned above Text Embedding 3 Small for both English and non-English tasks, prioritising embedding quality over the smaller model's lower cost.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/text-embedding-3-large","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/guides/embeddings","title":"OpenAI embeddings guide"}],"display_name":"Text Embedding 3 Large","context_window":8192,"reasoning":null,"output_modality":"embedding","output_modalities":["embedding"],"endpoints":["/v1/embeddings"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.13,"output_per_mtok_usd":0}},{"id":"openai/gpt-5.5","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":44.7,"artificial_analysis_coding_index":74.9,"artificial_analysis_agentic_index":36.4,"artificial_analysis_economics_index":47.8,"artificial_analysis_engineering_index":39.9,"artificial_analysis_intelligence_index":38.4,"artificial_analysis_strategy_and_ops_index":45.2,"artificial_analysis_finance_and_accounting_index":41.5,"artificial_analysis_healthcare_and_medical_index":38.5}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.457831325301205,"lcr":0.843333333333333,"gpqa":0.935353535353535,"tau2":0.93859649122807,"aime25":null,"critpt":0.271428571428571,"gdpval":1350.71,"ifbench":0.758503401360544,"mmmuPro":0.798843930635838,"scicode":0.55787037037037,"apexAgents":0.376843657817109,"hleDiamond":null,"itBenchSre":0.458097928436911,"licenseUrl":null,"parameters":null,"tauBanking":0.389690721649485,"licenseName":null,"mlcrOverall":0.288888888888889,"omniscience":20.516666666666666,"analystAgent":0.5,"frontierCode":null,"cacheHitPrice":0.5,"gdpPdfAllPass":0.212,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.842696629213483,"terminalBench40":0.146464646464646,"enterpriseOpsGym":0.4664279319606088,"gdpvalNormalized":0.42535500000000004,"intelligenceIndex":38.3556437590607,"terminalbenchHard":0.606060606060606,"briefcaseTotalCost":468.3792039498885,"price1mInputTokens":5,"contextWindowTokens":922000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.5795,"price1mOutputTokens":30,"pricePer1k1mpImages":5.12,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":17.5,"price1mBlended0To3To1":11.25,"price1mBlended7To2To1":4.35,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":5.247524752475248,"price1mBlended100To1To1":0.8333333333333334,"automationBenchPartialScore":0.4730056587293313,"intelligenceIndexTimePerTask":270.3093755493311,"omniscienceHallucinationRate":0.890210067380103,"inferenceParametersActiveBillions":null},"name":"GPT-5.5 (Xhigh)","slug":"gpt-5-5","source":"https://artificialanalysis.ai/models/gpt-5-5","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.5/benchmarks"},"created":0,"released_at":"2026-04-23","release_source":"https://openai.com/index/introducing-gpt-5-5/","description":"GPT-5.5 is an OpenAI flagship model for the most complex professional work, positioned as a new class of intelligence for coding and knowledge work. It builds on GPT-5.4 with stronger reasoning, higher reliability, and better token efficiency on hard tasks, with text and image input and tool use.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.5","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.5","title":"OpenAI model documentation"}],"display_name":"GPT-5.5","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":30,"cached_input_per_mtok_usd":0.5,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":10,"output_per_mtok_usd":45,"cached_input_per_mtok_usd":1}]}},{"id":"openai/gpt-5.5-pro","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":null,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":null,"lcr":null,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.305714285714286,"gdpval":null,"ifbench":null,"mmmuPro":null,"scicode":null,"sizeClass":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":null,"hostModelCount":0,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":null,"terminalbenchHard":null,"price1mInputTokens":null,"contextWindowTokens":922000,"knowledgeCutoffDate":null,"price1mOutputTokens":null,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":null,"price1mBlended0To3To1":null,"price1mBlended7To2To1":null,"cacheHitDiscountPercent":null,"price1mBlended0To100To1":null,"price1mBlended100To1To1":null,"inferenceParametersActiveBillions":null},"name":"GPT-5.5 Pro (Xhigh)","slug":"gpt-5-5-pro","source":"https://artificialanalysis.ai/models/gpt-5-5-pro","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.5-pro/benchmarks"},"created":0,"released_at":"2026-04-23","release_source":"https://openai.com/index/introducing-gpt-5-5/","description":"GPT-5.5 Pro is the high-compute version of OpenAI's GPT-5.5, aimed at deep reasoning and accuracy on complex, high-stakes workloads. It thinks harder to produce smarter, more precise answers than base GPT-5.5, runs only through the Responses and Batch endpoints, and may take minutes to respond, so background mode is recommended.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.5-pro","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.5-pro","title":"OpenAI model documentation"}],"display_name":"GPT-5.5 Pro","context_window":1050000,"reasoning":{"kind":"effort","levels":["medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":30,"output_per_mtok_usd":180,"cached_input_per_mtok_usd":3,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":60,"output_per_mtok_usd":270}]}},{"id":"openai/gpt-5.4","providers":["azure","azure-foundry","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":71.1,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":39,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.437442075996293,"lcr":0.82,"gpqa":0.92020202020202,"tau2":0.871345029239766,"aime25":null,"critpt":0.2342857143,"gdpval":1246.23,"ifbench":0.739455782312925,"mmmuPro":0.784393063583815,"scicode":null,"apexAgents":0.33259587020649,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.395876288659794,"licenseName":null,"omniscience":5.783333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.25,"livecodebench":null,"hostModelCount":3,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.782771535580524,"terminalBench40":null,"gdpvalNormalized":0.37311500000000003,"intelligenceIndex":38.9755886799637,"terminalbenchHard":0.575757575757576,"price1mInputTokens":2.5,"contextWindowTokens":1050000,"omniscienceAccuracy":0.5085,"price1mOutputTokens":15,"pricePer1k1mpImages":2.56,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":8.75,"price1mBlended0To3To1":5.625,"price1mBlended7To2To1":2.175,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.623762376237624,"price1mBlended100To1To1":0.4166666666666667,"omniscienceHallucinationRate":0.916920990166158,"inferenceParametersActiveBillions":null},"name":"GPT-5.4 (Xhigh)","slug":"gpt-5-4","source":"https://artificialanalysis.ai/models/gpt-5-4","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.4/benchmarks"},"created":0,"released_at":"2026-03-05","release_source":"https://openai.com/index/introducing-gpt-5-4/","description":"GPT-5.4 is an OpenAI flagship model for complex professional work, positioned as a more affordable option for coding and knowledge work. It unified the Codex and GPT lines into one system, supports text and image input with configurable reasoning effort, and anchors a family of mini, nano, and Pro tiers.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.4","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.4","title":"OpenAI model documentation"}],"display_name":"GPT-5.4","context_window":1050000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2.5,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.25,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":5,"output_per_mtok_usd":22.5,"cached_input_per_mtok_usd":0.5}]}},{"id":"openai/gpt-5.4-mini","providers":["azure","azure-foundry","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":26.2,"artificial_analysis_coding_index":56.1,"artificial_analysis_agentic_index":17.9,"artificial_analysis_economics_index":29.7,"artificial_analysis_engineering_index":24.4,"artificial_analysis_intelligence_index":24.1,"artificial_analysis_strategy_and_ops_index":25.3,"artificial_analysis_finance_and_accounting_index":24.2,"artificial_analysis_healthcare_and_medical_index":22.1}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.281278962001854,"lcr":0.77,"gpqa":0.874747474747475,"tau2":0.833333333333333,"aime25":null,"critpt":0.1,"gdpval":1011.97,"ifbench":0.73265306122449,"mmmuPro":0.732947976878613,"scicode":0.520833333333333,"apexAgents":0.281710914454277,"hleDiamond":null,"itBenchSre":0.352165725047081,"licenseUrl":null,"parameters":null,"tauBanking":0.255670103092783,"licenseName":null,"mlcrOverall":0.2,"omniscience":-18.916666666666668,"analystAgent":0.15,"frontierCode":null,"cacheHitPrice":0.075,"gdpPdfAllPass":0.134,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.591760299625468,"terminalBench40":0.0202020202020202,"enterpriseOpsGym":0.3461653237839451,"gdpvalNormalized":0.255985,"intelligenceIndex":24.0682169231964,"terminalbenchHard":0.522727272727273,"briefcaseTotalCost":34.05336575223885,"price1mInputTokens":0.75,"contextWindowTokens":400000,"omniscienceAccuracy":0.37466666666666665,"price1mOutputTokens":4.5,"pricePer1k1mpImages":1.24,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":2.625,"price1mBlended0To3To1":1.6875,"price1mBlended7To2To1":0.6525000000000001,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.7871287128712872,"price1mBlended100To1To1":0.125,"automationBenchPartialScore":0.26204887225711754,"intelligenceIndexTimePerTask":189.4909715645488,"omniscienceHallucinationRate":0.9016524520255863,"inferenceParametersActiveBillions":null},"name":"GPT-5.4 mini (Xhigh)","slug":"gpt-5-4-mini","source":"https://artificialanalysis.ai/models/gpt-5-4-mini","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.4-mini/benchmarks"},"created":0,"released_at":"2026-03-17","release_source":"https://openai.com/index/introducing-gpt-5-4-mini-and-nano/","description":"GPT-5.4 mini is the faster, more efficient tier of OpenAI's GPT-5.4 family, designed for high-volume workloads. OpenAI calls it its strongest mini model yet for coding, computer use, and subagents, keeping text and image input and strong reasoning while sitting between GPT-5.4 and the lighter nano.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.4-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.4-mini","title":"OpenAI model documentation"}],"display_name":"GPT-5.4 mini","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.75,"output_per_mtok_usd":4.5,"cached_input_per_mtok_usd":0.075}},{"id":"openai/gpt-5.4-nano","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":22.1,"artificial_analysis_coding_index":56.1,"artificial_analysis_agentic_index":16,"artificial_analysis_economics_index":30.3,"artificial_analysis_engineering_index":22.7,"artificial_analysis_intelligence_index":20.7,"artificial_analysis_strategy_and_ops_index":15.8,"artificial_analysis_finance_and_accounting_index":20.7,"artificial_analysis_healthcare_and_medical_index":16.8}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.282669138090825,"lcr":0.766666666666667,"gpqa":0.817171717171717,"tau2":0.760233918128655,"aime25":null,"critpt":0.0925306122448979,"gdpval":947.98,"ifbench":0.759183673469388,"mmmuPro":0.653757225433526,"scicode":0.472222222222222,"apexAgents":0.249262536873156,"hleDiamond":null,"itBenchSre":0.243879472693032,"licenseUrl":null,"parameters":null,"tauBanking":0.274226804123711,"licenseName":null,"mlcrOverall":0.161111111111111,"omniscience":-29.466666666666665,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.02,"gdpPdfAllPass":0.078,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.606741573033708,"terminalBench40":0.00505050505050505,"enterpriseOpsGym":0.3148313936138466,"gdpvalNormalized":0.22399000000000002,"intelligenceIndex":20.7197352504557,"terminalbenchHard":0.424242424242424,"briefcaseTotalCost":53.51295909768926,"price1mInputTokens":0.2,"contextWindowTokens":400000,"omniscienceAccuracy":0.25666666666666665,"price1mOutputTokens":1.25,"pricePer1k1mpImages":0.5,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.725,"price1mBlended0To3To1":0.4625,"price1mBlended7To2To1":0.179,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.2103960396039604,"price1mBlended100To1To1":0.03382352941176471,"automationBenchPartialScore":0.05960613074868808,"intelligenceIndexTimePerTask":410.47843920298516,"omniscienceHallucinationRate":0.7417040358744394,"inferenceParametersActiveBillions":null},"name":"GPT-5.4 nano (Xhigh)","slug":"gpt-5-4-nano","source":"https://artificialanalysis.ai/models/gpt-5-4-nano","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.4-nano/benchmarks"},"created":0,"released_at":"2026-03-17","release_source":"https://openai.com/index/introducing-gpt-5-4-mini-and-nano/","description":"GPT-5.4 nano is the lightest, most cost-efficient tier of OpenAI's GPT-5.4 family, optimised for speed-critical and high-volume tasks. It supports text and image input and targets low-latency work, trading the depth of GPT-5.4 and GPT-5.4 mini for throughput and price.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.4-nano","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.4-nano","title":"OpenAI model documentation"}],"display_name":"GPT-5.4 nano","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.2,"output_per_mtok_usd":1.25,"cached_input_per_mtok_usd":0.02}},{"id":"openai/gpt-5.4-pro","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":null,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":null,"lcr":null,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.3,"gdpval":null,"ifbench":null,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":null,"terminalbenchHard":null,"price1mInputTokens":30,"contextWindowTokens":1050000,"price1mOutputTokens":180,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":105,"price1mBlended0To3To1":67.5,"price1mBlended7To2To1":45,"cacheHitDiscountPercent":null,"price1mBlended0To100To1":31.485148514851485,"price1mBlended100To1To1":31.470588235294116,"inferenceParametersActiveBillions":null},"name":"GPT-5.4 Pro (Xhigh)","slug":"gpt-5-4-pro","source":"https://artificialanalysis.ai/models/gpt-5-4-pro","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.4-pro/benchmarks"},"created":0,"released_at":"2026-03-05","release_source":"https://openai.com/index/introducing-gpt-5-4/","description":"GPT-5.4 Pro is the high-compute version of OpenAI's GPT-5.4, built to tackle tough problems on complex, high-stakes tasks. It uses more compute to think harder and give consistently more precise answers than base GPT-5.4, and it is available only through the Responses API.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.4-pro","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.4-pro","title":"OpenAI model documentation"}],"display_name":"GPT-5.4 Pro","context_window":1050000,"reasoning":{"kind":"effort","levels":["medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":30,"output_per_mtok_usd":180,"cached_input_per_mtok_usd":3,"context_price_tiers":[{"above_input_tokens":272000,"input_per_mtok_usd":60,"output_per_mtok_usd":270}]}},{"id":"openai/gpt-5.3-codex","providers":["azure"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":32.5,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.424930491195551,"lcr":0.833333333333333,"gpqa":0.915151515151515,"tau2":0.859649122807018,"aime25":null,"critpt":0.168571428571429,"gdpval":null,"ifbench":0.753741496598639,"mmmuPro":0.784971098265896,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":10.866666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.175,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":32.5028174368983,"terminalbenchHard":0.53030303030303,"briefcaseTotalCost":189.6501659170993,"price1mInputTokens":1.75,"contextWindowTokens":400000,"omniscienceAccuracy":0.5288333333333334,"price1mOutputTokens":14,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":7.875,"price1mBlended0To3To1":4.8125,"price1mBlended7To2To1":1.8725,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.8712871287128714,"price1mBlended100To1To1":0.32598039215686275,"omniscienceHallucinationRate":0.8917580474000707,"inferenceParametersActiveBillions":null},"name":"GPT-5.3 Codex (Xhigh)","slug":"gpt-5-3-codex","source":"https://artificialanalysis.ai/models/gpt-5-3-codex","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.3-codex/benchmarks"},"created":0,"released_at":"2026-02-05","release_source":"https://openai.com/index/introducing-gpt-5-3-codex/","description":"GPT-5.3 Codex is OpenAI's agentic coding model, optimised for autonomous software engineering in Codex and similar environments. It combines the engineering strengths of GPT-5.2 Codex with the broader reasoning and professional knowledge of GPT-5.2, and supports reasoning effort from low up to xhigh.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.3-codex","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.3-codex","title":"OpenAI model documentation"}],"display_name":"GPT-5.3 Codex","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.75,"output_per_mtok_usd":14,"cached_input_per_mtok_usd":0.175}},{"id":"openai/gpt-5.1","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":49.4,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":24.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.28498609823911,"lcr":0.8,"gpqa":0.872727272727273,"tau2":0.818713450292398,"aime25":0.94,"critpt":0.0485714285714286,"gdpval":826.29,"ifbench":0.728571428571429,"mmmuPro":0.754913294797688,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.158762886597938,"licenseName":null,"omniscience":5.4,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.125,"livecodebench":0.867724867724868,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.524344569288389,"terminalBench40":null,"gdpvalNormalized":0.16314499999999998,"intelligenceIndex":24.7357540558308,"terminalbenchHard":0.454545454545455,"price1mInputTokens":1.25,"contextWindowTokens":400000,"omniscienceAccuracy":0.37716666666666665,"price1mOutputTokens":10,"pricePer1k1mpImages":0.788,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5.625,"price1mBlended0To3To1":3.4375,"price1mBlended7To2To1":1.3375,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.3366336633663367,"price1mBlended100To1To1":0.23284313725490197,"omniscienceHallucinationRate":0.5188654000535189,"inferenceParametersActiveBillions":null},"name":"GPT-5.1 (High)","slug":"gpt-5-1","source":"https://artificialanalysis.ai/models/gpt-5-1","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.1/benchmarks"},"created":0,"released_at":"2025-11-12","release_source":"https://openai.com/index/gpt-5-1/","description":"GPT-5.1 is OpenAI's GPT-5 series flagship for coding and agentic tasks, with stronger reasoning, better instruction adherence, and a more natural style than GPT-5. Its distinguishing feature is adaptive, configurable reasoning effort, including a none setting for fast non-reasoning responses.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.1","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.1","title":"OpenAI model documentation"}],"display_name":"GPT-5.1","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.125}},{"id":"openai/gpt-5.2","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":30.4,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.376737720111214,"lcr":0.826666666666667,"gpqa":0.903030303030303,"tau2":0.847953216374269,"aime25":0.99,"critpt":0.115510204081633,"gdpval":null,"ifbench":0.754421768707483,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-0.8833333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.175,"livecodebench":0.888888888888889,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":30.448181916806,"terminalbenchHard":0.46969696969697,"price1mInputTokens":1.75,"contextWindowTokens":400000,"omniscienceAccuracy":0.44316666666666665,"price1mOutputTokens":14,"pricePer1k1mpImages":2.15,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":7.875,"price1mBlended0To3To1":4.8125,"price1mBlended7To2To1":1.8725,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.8712871287128714,"price1mBlended100To1To1":0.32598039215686275,"omniscienceHallucinationRate":0.8117330140676444,"inferenceParametersActiveBillions":null},"name":"GPT-5.2 (Xhigh)","slug":"gpt-5-2","source":"https://artificialanalysis.ai/models/gpt-5-2","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.2/benchmarks"},"created":0,"released_at":"2025-12-11","release_source":"https://openai.com/index/introducing-gpt-5-2/","description":"GPT-5.2 is a previous OpenAI flagship for complex professional work, following GPT-5.1 in the GPT-5 series. It improves agentic and long-context performance over GPT-5.1 and keeps configurable reasoning effort from none through xhigh, allocating compute adaptively so simple requests return quickly.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.2","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.2","title":"OpenAI model documentation"}],"display_name":"GPT-5.2","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.75,"output_per_mtok_usd":14,"cached_input_per_mtok_usd":0.175}},{"id":"openai/gpt-5.2-codex","providers":["azure"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":28.5,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.357275254865616,"lcr":0.823333333333333,"gpqa":0.898989898989899,"tau2":0.921052631578947,"aime25":null,"critpt":0.0866938775510204,"gdpval":null,"ifbench":0.776190476190476,"mmmuPro":0.763005780346821,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-2.1666666666666665,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.175,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":28.5016971971953,"terminalbenchHard":0.371212121212121,"price1mInputTokens":1.75,"contextWindowTokens":400000,"omniscienceAccuracy":0.4106666666666667,"price1mOutputTokens":14,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":7.875,"price1mBlended0To3To1":4.8125,"price1mBlended7To2To1":1.8725,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.8712871287128714,"price1mBlended100To1To1":0.32598039215686275,"omniscienceHallucinationRate":0.7335972850678733,"inferenceParametersActiveBillions":null},"name":"GPT-5.2 Codex (Xhigh)","slug":"gpt-5-2-codex","source":"https://artificialanalysis.ai/models/gpt-5-2-codex","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.2-codex/benchmarks"},"created":0,"released_at":"2025-12-18","release_source":"https://openai.com/index/introducing-gpt-5-2-codex/","description":"GPT-5.2 Codex is the software engineering version of OpenAI's GPT-5.2, an upgrade over GPT-5.1 Codex for coding workflows. It is built for both interactive development sessions and long, independent execution of complex engineering tasks in Codex-style agentic environments.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.2-codex","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.2-codex","title":"OpenAI model documentation"}],"display_name":"GPT-5.2 Codex","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.75,"output_per_mtok_usd":14,"cached_input_per_mtok_usd":0.175}},{"id":"openai/gpt-5.1-codex","providers":["azure"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":23.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T15:18:54.161Z"},"data":{"hle":0.257182576459685,"lcr":0.693333333333333,"gpqa":0.85959595959596,"tau2":0.830409356725146,"aime25":0.956666666666667,"critpt":0.0571428571428571,"gdpval":null,"ifbench":0.7,"mmmuPro":0.72485549132948,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-6.5,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.125,"livecodebench":0.848677248677249,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":23.6972790075055,"terminalbenchHard":0.348484848484849,"price1mInputTokens":1.25,"contextWindowTokens":400000,"omniscienceAccuracy":0.399,"price1mOutputTokens":10,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5.625,"price1mBlended0To3To1":3.4375,"price1mBlended7To2To1":1.3375,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.3366336633663367,"price1mBlended100To1To1":0.23284313725490197,"omniscienceHallucinationRate":0.7720465890183028,"inferenceParametersActiveBillions":null},"name":"GPT-5.1 Codex (High)","slug":"gpt-5-1-codex","source":"https://artificialanalysis.ai/models/gpt-5-1-codex","fetchedAt":"2026-10-21T15:18:54.161Z","pageRecords":[]}],"checkedAt":"2026-10-21T15:18:54.161Z","fetchedAt":"2026-10-21T15:18:54.161Z","nextCheckAt":"2026-10-28T15:18:54.161Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.1-codex/benchmarks"},"created":0,"released_at":"2025-11-13","release_source":"https://openai.com/index/gpt-5-1-for-developers/","description":"GPT-5.1 Codex is a specialised version of OpenAI's GPT-5.1 optimised for software engineering and coding workflows. It is designed for interactive development sessions as well as long, independent runs on complex engineering tasks, and is the base for the smaller GPT-5.1 Codex mini.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.1-codex","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.1-codex","title":"OpenAI model documentation"}],"display_name":"GPT-5.1 Codex","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.125}},{"id":"openai/gpt-5.1-codex-mini","providers":["azure"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":20.4,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.184893419833179,"lcr":0.666666666666667,"gpqa":0.813131313131313,"tau2":0.628654970760234,"aime25":0.916666666666667,"critpt":0,"gdpval":null,"ifbench":0.67891156462585,"mmmuPro":0.690173410404624,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-16.416666666666668,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.025,"livecodebench":0.835978835978836,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":20.3781988302265,"terminalbenchHard":0.333333333333333,"price1mInputTokens":0.25,"contextWindowTokens":400000,"omniscienceAccuracy":0.232,"price1mOutputTokens":2,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1.125,"price1mBlended0To3To1":0.6875,"price1mBlended7To2To1":0.26749999999999996,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.26732673267326734,"price1mBlended100To1To1":0.04656862745098039,"omniscienceHallucinationRate":0.5158420138888888,"inferenceParametersActiveBillions":null},"name":"GPT-5.1 Codex mini (High)","slug":"gpt-5-1-codex-mini","source":"https://artificialanalysis.ai/models/gpt-5-1-codex-mini","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-5.1-codex-mini/benchmarks"},"created":0,"released_at":"2025-11-13","release_source":"https://openai.com/index/gpt-5-1-for-developers/","description":"GPT-5.1 Codex mini is the smaller, faster version of OpenAI's GPT-5.1 Codex, aimed at the same agentic software engineering workflows. It trades some of the full model's capability for lower latency, making it suited to lighter coding tasks and higher-volume Codex usage.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5.1-codex-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5.1-codex-mini","title":"OpenAI model documentation"}],"display_name":"GPT-5.1 Codex mini","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.25,"output_per_mtok_usd":2,"cached_input_per_mtok_usd":0.025}},{"id":"openai/chat-latest","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T19:05:55.416Z","nextCheckAt":"2026-10-28T19:05:55.416Z","summary":true,"detailPath":"/v1/models/openai/chat-latest/benchmarks"},"created":0,"released_at":"2026-05-05","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT Chat Latest is OpenAI's rolling alias that points to the latest Instant model currently used in ChatGPT, with text and image input and tool support. Unlike a pinned snapshot, the underlying model is updated regularly, so OpenAI recommends GPT-6 Astra for production work that needs stability.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/chat-latest","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"}],"display_name":"GPT Chat Latest","context_window":400000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":30,"cached_input_per_mtok_usd":0.5}},{"id":"openai/gpt-5","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":37.8,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":23,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.28498609823911,"lcr":0.782,"gpqa":0.853535353535353,"tau2":0.847953216374269,"aime25":0.943333333333333,"critpt":0.0571428571428571,"gdpval":915.8,"ifbench":0.730612244897959,"mmmuPro":0.742196531791907,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.220618556701031,"licenseName":null,"omniscience":-8.733333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.125,"livecodebench":0.845502645502646,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.352059925093633,"terminalBench40":null,"gdpvalNormalized":0.20789999999999997,"intelligenceIndex":22.9827995216189,"terminalbenchHard":0.325757575757576,"price1mInputTokens":1.25,"contextWindowTokens":400000,"omniscienceAccuracy":0.4031666666666667,"price1mOutputTokens":10,"pricePer1k1mpImages":0.788,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5.625,"price1mBlended0To3To1":3.4375,"price1mBlended7To2To1":1.3375,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.3366336633663367,"price1mBlended100To1To1":0.23284313725490197,"omniscienceHallucinationRate":0.8218374755654845,"inferenceParametersActiveBillions":null},"name":"GPT-5 (High)","slug":"gpt-5","source":"https://artificialanalysis.ai/models/gpt-5","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-5/benchmarks"},"created":0,"released_at":"2025-08-07","release_source":"https://openai.com/index/introducing-gpt-5-for-developers/","description":"GPT-5 is OpenAI's frontier model of its generation, with major improvements in reasoning, code quality, and user experience over GPT-4.1. It is optimised for complex tasks that require step-by-step reasoning, instruction following, and accuracy, and heads a family with mini, nano, Pro, and Codex variants.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5","title":"OpenAI model documentation"}],"display_name":"GPT-5","context_window":400000,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.125}},{"id":"openai/gpt-5-pro","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T19:05:55.416Z","nextCheckAt":"2026-10-28T19:05:55.416Z","summary":true,"detailPath":"/v1/models/openai/gpt-5-pro/benchmarks"},"created":0,"released_at":"2025-08-07","release_source":"https://openai.com/index/introducing-gpt-5/","description":"GPT-5 Pro is the high-compute version of OpenAI's GPT-5, aimed at the hardest problems in reasoning, coding, and analysis. It uses extended thinking to give more accurate and precise answers than base GPT-5 and is available through the Responses API for tasks where quality outweighs latency.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5-pro","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5-pro","title":"OpenAI model documentation"}],"display_name":"GPT-5 Pro","context_window":400000,"reasoning":{"kind":"effort","levels":["high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":15,"output_per_mtok_usd":120}},{"id":"openai/gpt-5-codex","providers":["azure"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":24.9,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.278498609823911,"lcr":0.716666666666667,"gpqa":0.837373737373737,"tau2":0.868421052631579,"aime25":0.986666666666667,"critpt":0.0514285714285714,"gdpval":null,"ifbench":0.741496598639456,"mmmuPro":0.738150289017341,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-7.533333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.125,"livecodebench":0.84021164021164,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":24.8818938471151,"terminalbenchHard":0.378787878787879,"price1mInputTokens":1.25,"contextWindowTokens":400000,"omniscienceAccuracy":0.387,"price1mOutputTokens":10,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5.625,"price1mBlended0To3To1":3.4375,"price1mBlended7To2To1":1.3375,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.3366336633663367,"price1mBlended100To1To1":0.23284313725490197,"omniscienceHallucinationRate":0.7542142468733007,"inferenceParametersActiveBillions":null},"name":"GPT-5 Codex (High)","slug":"gpt-5-codex","source":"https://artificialanalysis.ai/models/gpt-5-codex","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-5-codex/benchmarks"},"created":0,"released_at":"2025-09-15","release_source":"https://openai.com/index/introducing-upgrades-to-codex/","description":"GPT-5 Codex is a version of OpenAI's GPT-5 optimised for agentic coding in Codex or similar environments. It shares GPT-5's text and image input and reasoning, but is tuned for autonomous, tool-driven software engineering and is available only through the Responses API.","description_sources":[{"url":"https://developers.openai.com/api/docs/models/gpt-5-codex","title":"OpenAI model documentation"},{"url":"https://developers.openai.com/api/docs/changelog","title":"OpenAI API changelog"}],"display_name":"GPT-5 Codex","context_window":400000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.125}},{"id":"openai/gpt-5-mini","providers":["azure","azure-foundry","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":20,"artificial_analysis_coding_index":15.6,"artificial_analysis_agentic_index":6.8,"artificial_analysis_economics_index":27.7,"artificial_analysis_engineering_index":20.7,"artificial_analysis_intelligence_index":16.8,"artificial_analysis_strategy_and_ops_index":12.4,"artificial_analysis_finance_and_accounting_index":17.2,"artificial_analysis_healthcare_and_medical_index":16.9}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.214550509731233,"lcr":0.723333333333333,"gpqa":0.828282828282828,"tau2":0.684210526315789,"aime25":0.906666666666667,"critpt":0,"gdpval":763.51,"ifbench":0.754421768707483,"mmmuPro":0.700578034682081,"scicode":0.390046296296296,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.154639175257732,"licenseName":null,"mlcrOverall":0.0888888888888889,"omniscience":-17.316666666666666,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.025,"gdpPdfAllPass":0.084,"livecodebench":0.838095238095238,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.0374531835205993,"terminalBench40":0,"gdpvalNormalized":0.13175499999999998,"intelligenceIndex":16.7730881042515,"terminalbenchHard":0.333333333333333,"briefcaseTotalCost":7.523888653950295,"price1mInputTokens":0.25,"contextWindowTokens":400000,"omniscienceAccuracy":0.24966666666666668,"price1mOutputTokens":2,"pricePer1k1mpImages":0.41,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1.125,"price1mBlended0To3To1":0.6875,"price1mBlended7To2To1":0.26749999999999996,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.26732673267326734,"price1mBlended100To1To1":0.04656862745098039,"automationBenchPartialScore":0.06487466331916088,"intelligenceIndexTimePerTask":147.14483409941337,"omniscienceHallucinationRate":0.5635273211905819,"inferenceParametersActiveBillions":null},"name":"GPT-5 mini (High)","slug":"gpt-5-mini","source":"https://artificialanalysis.ai/models/gpt-5-mini","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-5-mini/benchmarks"},"created":0,"released_at":"2025-08-07","release_source":"https://openai.com/index/introducing-gpt-5-for-developers/","description":"GPT-5 mini is the compact tier of OpenAI's GPT-5 family, designed for lighter-weight reasoning tasks at lower latency and cost. It keeps GPT-5's instruction-following and safety tuning while sitting between the full model and the smaller GPT-5 nano.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5-mini","title":"OpenAI model documentation"}],"display_name":"GPT-5 mini","context_window":400000,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.25,"output_per_mtok_usd":2,"cached_input_per_mtok_usd":0.025}},{"id":"openai/gpt-5-nano","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":13,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0949953660797034,"lcr":0.45,"gpqa":0.675757575757576,"tau2":0.365497076023392,"aime25":0.836666666666667,"critpt":0,"gdpval":null,"ifbench":0.675510204081633,"mmmuPro":0.609826589595376,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-28.666666666666668,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.005,"livecodebench":0.789417989417989,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":12.992236382483,"terminalbenchHard":0.121212121212121,"price1mInputTokens":0.05,"contextWindowTokens":400000,"omniscienceAccuracy":0.19083333333333333,"price1mOutputTokens":0.4,"pricePer1k1mpImages":0.13,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.225,"price1mBlended0To3To1":0.1375,"price1mBlended7To2To1":0.053500000000000006,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.05346534653465347,"price1mBlended100To1To1":0.009313725490196078,"omniscienceHallucinationRate":0.5901132852729145,"inferenceParametersActiveBillions":null},"name":"GPT-5 nano (High)","slug":"gpt-5-nano","source":"https://artificialanalysis.ai/models/gpt-5-nano","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-5-nano/benchmarks"},"created":0,"released_at":"2025-08-07","release_source":"https://openai.com/index/introducing-gpt-5-for-developers/","description":"GPT-5 nano is the smallest and fastest tier of OpenAI's GPT-5 family, optimised for developer tools, rapid interactions, and ultra-low-latency environments. It offers less reasoning depth than GPT-5 and GPT-5 mini, trading capability for speed and throughput on high-volume tasks.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-5-nano","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-5-nano","title":"OpenAI model documentation"}],"display_name":"GPT-5 nano","context_window":400000,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.05,"output_per_mtok_usd":0.4,"cached_input_per_mtok_usd":0.005}},{"id":"openai/gpt-4.1","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":12.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0418,"lcr":0.683333333333333,"gpqa":0.665656565656566,"tau2":0.470760233918129,"aime25":0.346666666666667,"critpt":0,"gdpval":null,"suffix":null,"ifbench":0.429931972789116,"mmmuPro":0.612138728323699,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-39.63333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.5,"livecodebench":0.457142857142857,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":12.6913226077818,"terminalbenchHard":0.136363636363636,"price1mInputTokens":2,"contextWindowTokens":1000000,"omniscienceAccuracy":0.2776666666666667,"price1mOutputTokens":8,"pricePer1k1mpImages":1.53,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5,"price1mBlended0To3To1":3.5,"price1mBlended7To2To1":1.55,"cacheHitDiscountPercent":0.75,"price1mBlended0To100To1":2.0594059405940595,"price1mBlended100To1To1":0.5882352941176471,"omniscienceHallucinationRate":0.9330872173511767,"inferenceParametersActiveBillions":null},"name":"GPT-4.1","slug":"gpt-4-1","source":"https://artificialanalysis.ai/models/gpt-4-1","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-4.1/benchmarks"},"created":0,"released_at":"2025-04-14","release_source":"https://openai.com/index/gpt-4-1/","description":"GPT-4.1 is OpenAI's flagship of the GPT-4.1 series, optimised for advanced instruction following, real-world software engineering, and long-context reasoning. It improves on GPT-4o across coding and instruction tasks and heads a family with mini and nano tiers for lower latency and cost.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-4.1","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-4.1","title":"OpenAI model documentation"}],"display_name":"GPT-4.1","context_window":1000000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":8,"cached_input_per_mtok_usd":0.5}},{"id":"openai/gpt-4.1-mini","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":20.2,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":10.2,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0502,"lcr":0.44,"gpqa":0.663636363636364,"tau2":0.529239766081871,"aime25":0.463333333333333,"critpt":0,"gdpval":270.29,"suffix":null,"ifbench":0.382993197278912,"mmmuPro":0.58728323699422,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.0536082474226804,"licenseName":null,"omniscience":-53.56666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.1,"livecodebench":0.482539682539683,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.101123595505618,"terminalBench40":null,"gdpvalNormalized":0,"intelligenceIndex":10.1588744766829,"terminalbenchHard":0.0757575757575758,"price1mInputTokens":0.4,"contextWindowTokens":1000000,"omniscienceAccuracy":0.20316666666666666,"price1mOutputTokens":1.6,"pricePer1k1mpImages":0.6636,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1,"price1mBlended0To3To1":0.7000000000000001,"price1mBlended7To2To1":0.31,"cacheHitDiscountPercent":0.75,"price1mBlended0To100To1":0.4118811881188119,"price1mBlended100To1To1":0.11764705882352941,"omniscienceHallucinationRate":0.9272118803597573,"inferenceParametersActiveBillions":null},"name":"GPT-4.1 mini","slug":"gpt-4-1-mini","source":"https://artificialanalysis.ai/models/gpt-4-1-mini","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-4.1-mini/benchmarks"},"created":0,"released_at":"2025-04-14","release_source":"https://openai.com/index/gpt-4-1/","description":"GPT-4.1 mini is the mid-sized tier of OpenAI's GPT-4.1 series, delivering performance competitive with GPT-4o at substantially lower latency and cost. It keeps the family's long-context and instruction-following strengths while sitting between the flagship GPT-4.1 and the faster GPT-4.1 nano.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-4.1-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-4.1-mini","title":"OpenAI model documentation"}],"display_name":"GPT-4.1 mini","context_window":1000000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.4,"output_per_mtok_usd":1.6,"cached_input_per_mtok_usd":0.1}},{"id":"openai/gpt-4.1-nano","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":11.1,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":7.8,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0375347544022243,"lcr":0.203333333333333,"gpqa":0.512121212121212,"tau2":0.172514619883041,"aime25":0.24,"critpt":0,"gdpval":-213.03,"suffix":null,"ifbench":0.320408163265306,"mmmuPro":0.401156069364162,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.0350515463917526,"licenseName":null,"omniscience":-57.583333333333336,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.025,"livecodebench":0.325925925925926,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.0374531835205993,"terminalBench40":null,"gdpvalNormalized":0,"intelligenceIndex":7.81796186868883,"terminalbenchHard":0.0378787878787879,"price1mInputTokens":0.1,"contextWindowTokens":1000000,"omniscienceAccuracy":0.13683333333333333,"price1mOutputTokens":0.4,"pricePer1k1mpImages":0.2519,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.25,"price1mBlended0To3To1":0.17500000000000002,"price1mBlended7To2To1":0.0775,"cacheHitDiscountPercent":0.75,"price1mBlended0To100To1":0.10297029702970298,"price1mBlended100To1To1":0.029411764705882353,"omniscienceHallucinationRate":0.8256420158331724,"inferenceParametersActiveBillions":null},"name":"GPT-4.1 nano","slug":"gpt-4-1-nano","source":"https://artificialanalysis.ai/models/gpt-4-1-nano","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-4.1-nano/benchmarks"},"created":0,"released_at":"2025-04-14","release_source":"https://openai.com/index/gpt-4-1/","description":"GPT-4.1 nano is the fastest and cheapest tier of OpenAI's GPT-4.1 series, built for tasks that demand low latency. It suits classification, autocompletion, and other high-volume work, trading the flagship's depth for speed while retaining the family's long-context support.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-4.1-nano","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-4.1-nano","title":"OpenAI model documentation"}],"display_name":"GPT-4.1 nano","context_window":1000000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0.4,"cached_input_per_mtok_usd":0.025}},{"id":"openai/gpt-4o","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":8.4,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0242,"lcr":0.493333333333333,"gpqa":0.543434343434343,"tau2":0.251461988304094,"aime25":0.06,"critpt":0,"gdpval":-16.24,"suffix":null,"ifbench":0.342857142857143,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-10.516666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":1.5,"livecodebench":0.308994708994709,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"gdpvalNormalized":0,"intelligenceIndex":8.4416392330287,"terminalbenchHard":0.0833333333333333,"price1mInputTokens":2.5,"contextWindowTokens":128000,"omniscienceAccuracy":0.19866666666666666,"price1mOutputTokens":10,"pricePer1k1mpImages":1.9125,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":6.25,"price1mBlended0To3To1":4.375,"price1mBlended7To2To1":2.55,"cacheHitDiscountPercent":0.4,"price1mBlended0To100To1":2.5742574257425743,"price1mBlended100To1To1":1.5931372549019607,"omniscienceHallucinationRate":0.37915973377703827,"inferenceParametersActiveBillions":null},"name":"GPT-4o (Nov '24)","slug":"gpt-4o","source":"https://artificialanalysis.ai/models/gpt-4o","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-4o/benchmarks"},"created":0,"released_at":"2024-05-13","release_source":"https://openai.com/index/hello-gpt-4o/","description":"GPT-4o is OpenAI's omni model of the GPT-4 generation, accepting text and image input with text output for general chat, analysis, and coding. It matches the intelligence of GPT-4 Turbo while responding faster, and is the base for the smaller GPT-4o mini.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-4o","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-4o","title":"OpenAI model documentation"}],"display_name":"GPT-4o","context_window":128000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2.5,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":1.25}},{"id":"openai/gpt-4o-mini","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":11.4,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":6.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.042,"lcr":null,"gpqa":0.426262626262626,"tau2":null,"aime25":0.146666666666667,"critpt":null,"gdpval":-25.34,"suffix":null,"ifbench":0.30952380952381,"mmmuPro":0.415028901734104,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.0288659793814433,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.075,"livecodebench":0.233862433862434,"hostModelCount":3,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.0561797752808989,"terminalBench40":null,"gdpvalNormalized":0,"intelligenceIndex":6.66432562957107,"terminalbenchHard":null,"price1mInputTokens":0.15,"contextWindowTokens":128000,"price1mOutputTokens":0.6,"pricePer1k1mpImages":0.11475,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.375,"price1mBlended0To3To1":0.26249999999999996,"price1mBlended7To2To1":0.1425,"cacheHitDiscountPercent":0.5,"price1mBlended0To100To1":0.15445544554455445,"price1mBlended100To1To1":0.08088235294117647,"inferenceParametersActiveBillions":null},"name":"GPT-4o mini","slug":"gpt-4o-mini","source":"https://artificialanalysis.ai/models/gpt-4o-mini","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-4o-mini/benchmarks"},"created":0,"released_at":"2024-07-18","release_source":"https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/","description":"GPT-4o mini is OpenAI's small model of the GPT-4o generation, supporting text and image input for cost-efficient chat and everyday tasks. It was positioned as OpenAI's most capable small model at launch, offering much of GPT-4o's ability at far lower cost and latency.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-4o-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-4o-mini","title":"OpenAI model documentation"}],"display_name":"GPT-4o mini","context_window":128000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.15,"output_per_mtok_usd":0.6,"cached_input_per_mtok_usd":0.075}},{"id":"openai/gpt-4-turbo","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":21.5,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0313,"lcr":null,"gpqa":null,"tau2":null,"aime25":null,"critpt":null,"gdpval":null,"suffix":null,"ifbench":null,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":0.291005291005291,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":7.04351520157305,"terminalbenchHard":null,"price1mInputTokens":10,"contextWindowTokens":128000,"price1mOutputTokens":30,"pricePer1k1mpImages":7.65,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":20,"price1mBlended0To3To1":15,"price1mBlended7To2To1":12,"cacheHitDiscountPercent":null,"price1mBlended0To100To1":10.198019801980198,"price1mBlended100To1To1":10.196078431372548,"inferenceParametersActiveBillions":null},"name":"GPT-4 Turbo","slug":"gpt-4-turbo","source":"https://artificialanalysis.ai/models/gpt-4-turbo","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-4-turbo/benchmarks"},"created":0,"released_at":"2024-04-09","release_source":"https://developers.openai.com/api/docs/changelog","description":"GPT-4 Turbo is OpenAI's GPT-4 generation model with vision, an older flagship for chat, analysis, and coding over long inputs. Its vision requests support JSON mode and function calling, and it was later succeeded by the faster GPT-4o.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-4-turbo","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-4-turbo","title":"OpenAI model documentation"}],"display_name":"GPT-4 Turbo","context_window":128000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":10,"output_per_mtok_usd":30}},{"id":"openai/gpt-3.5-turbo","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["gpt-3-5-turbo: AA page HTTP 404"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":10.7,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":5.5,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-27T19:05:55.416Z"},"data":{"hle":null,"lcr":null,"gpqa":0.296969696969697,"tau2":null,"aime25":null,"critpt":null,"gdpval":null,"suffix":null,"ifbench":null,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"healthbench":null,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":null,"frontierCode20":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"critptAgenticHard":null,"intelligenceIndex":5.48558234391291,"terminalbenchHard":null,"price1mInputTokens":0.5,"contextWindowTokens":4096,"price1mOutputTokens":1.5,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1,"price1mBlended0To3To1":0.75,"price1mBlended7To2To1":0.6,"cacheHitDiscountPercent":null,"healthbenchProfessional":null,"price1mBlended0To100To1":0.5099009900990099,"price1mBlended100To1To1":0.5098039215686274,"inferenceParametersActiveBillions":null},"name":"GPT-3.5 Turbo","slug":"gpt-35-turbo","source":"https://artificialanalysis.ai/models/gpt-35-turbo","fetchedAt":"2026-10-27T19:05:55.416Z","pageRecords":[]}],"checkedAt":"2026-10-27T19:05:55.416Z","fetchedAt":"2026-09-11T10:50:14.034Z","nextCheckAt":"2026-10-28T19:05:55.416Z","summary":true,"detailPath":"/v1/models/openai/gpt-3.5-turbo/benchmarks"},"created":0,"released_at":"2023-03-01","release_source":"https://community.openai.com/t/introducing-chatgpt-and-whisper-apis/80485","description":"GPT-3.5 Turbo is OpenAI's earlier fast chat model, able to understand and generate natural language and code. It is optimised for chat and traditional completion tasks and served as the low-cost tier before GPT-4o mini took over that role.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-3.5-turbo","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-3.5-turbo","title":"OpenAI model documentation"}],"display_name":"GPT-3.5 Turbo","context_window":16000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.5,"output_per_mtok_usd":1.5}},{"id":"openai/o3","providers":["azure","openai"],"provider_details":[{"provider":"azure","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":20.2,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.2005,"lcr":0.746666666666667,"gpqa":0.827272727272727,"tau2":0.807017543859649,"aime25":0.883333333333333,"critpt":0.0114285714285714,"gdpval":null,"suffix":null,"ifbench":0.714285714285714,"mmmuPro":0.700578034682081,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-15.55,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.5,"livecodebench":0.808465608465608,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":20.200754789201,"terminalbenchHard":0.371212121212121,"price1mInputTokens":2,"contextWindowTokens":200000,"omniscienceAccuracy":0.38566666666666666,"price1mOutputTokens":8,"pricePer1k1mpImages":6.75,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5,"price1mBlended0To3To1":3.5,"price1mBlended7To2To1":1.55,"cacheHitDiscountPercent":0.75,"price1mBlended0To100To1":2.0594059405940595,"price1mBlended100To1To1":0.5882352941176471,"omniscienceHallucinationRate":0.8809007053716766,"inferenceParametersActiveBillions":null},"name":"o3","slug":"o3","source":"https://artificialanalysis.ai/models/o3","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/o3/benchmarks"},"created":0,"released_at":"2025-04-16","release_source":"https://openai.com/index/introducing-o3-and-o4-mini/","description":"OpenAI o3 is a reasoning model in OpenAI's o-series, well-rounded across math, science, coding, and visual reasoning, and strong at technical writing and instruction following. It can reason with images and call tools agentically, sitting above the compact o4-mini in the series.","description_sources":[{"url":"https://openrouter.ai/openai/o3","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/o3","title":"OpenAI model documentation"}],"display_name":"OpenAI o3","context_window":200000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":8,"cached_input_per_mtok_usd":0.5}},{"id":"openai/o3-mini","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":12.5,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0791,"lcr":null,"gpqa":0.748484848484848,"tau2":0.286549707602339,"aime25":null,"critpt":null,"gdpval":null,"suffix":null,"ifbench":null,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":null,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.55,"livecodebench":0.717460317460317,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":12.4707708274399,"terminalbenchHard":0.0681818181818182,"price1mInputTokens":1.1,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"price1mOutputTokens":4.4,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":2.75,"price1mBlended0To3To1":1.9250000000000003,"price1mBlended7To2To1":1.0450000000000002,"cacheHitDiscountPercent":0.5,"price1mBlended0To100To1":1.1326732673267328,"price1mBlended100To1To1":0.5931372549019609,"inferenceParametersActiveBillions":null},"name":"o3-mini","slug":"o3-mini","source":"https://artificialanalysis.ai/models/o3-mini","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/o3-mini/benchmarks"},"created":0,"released_at":"2025-01-31","release_source":"https://openai.com/index/openai-o3-mini/","description":"OpenAI o3-mini is the small, cost-efficient reasoning model in OpenAI's o-series, optimised for STEM tasks in science, mathematics, and coding. It exposes a reasoning_effort parameter so developers can trade speed for deeper thinking, and it is text-only unlike its successor o4-mini.","description_sources":[{"url":"https://openrouter.ai/openai/o3-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/o3-mini","title":"OpenAI model documentation"}],"display_name":"OpenAI o3-mini","context_window":200000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.1,"output_per_mtok_usd":4.4,"cached_input_per_mtok_usd":0.55}},{"id":"openai/o4-mini","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":16.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.165430954587581,"lcr":0.61,"gpqa":0.783838383838384,"tau2":0.555555555555556,"aime25":0.906666666666667,"critpt":0.00571428571428571,"gdpval":null,"ifbench":0.687074829931973,"mmmuPro":0.692485549132948,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-35.7,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.275,"livecodebench":0.859259259259259,"hostModelCount":3,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":16.653105214846,"terminalbenchHard":0.151515151515152,"price1mInputTokens":1.1,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.248,"price1mOutputTokens":4.4,"pricePer1k1mpImages":1.9371,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":2.75,"price1mBlended0To3To1":1.9250000000000003,"price1mBlended7To2To1":0.8525,"cacheHitDiscountPercent":0.75,"price1mBlended0To100To1":1.1326732673267328,"price1mBlended100To1To1":0.32352941176470595,"omniscienceHallucinationRate":0.8045212765957447,"inferenceParametersActiveBillions":null},"name":"o4-mini (High)","slug":"o4-mini","source":"https://artificialanalysis.ai/models/o4-mini","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/o4-mini/benchmarks"},"created":0,"released_at":"2025-04-16","release_source":"https://openai.com/index/introducing-o3-and-o4-mini/","description":"OpenAI o4-mini is a compact reasoning model in OpenAI's o-series, optimised for fast, cost-efficient performance. Unlike o3-mini it accepts images and supports tool use, retaining strong multimodal and agentic capabilities while sitting below the larger o3.","description_sources":[{"url":"https://openrouter.ai/openai/o4-mini","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/o4-mini","title":"OpenAI model documentation"}],"display_name":"OpenAI o4-mini","context_window":200000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.1,"output_per_mtok_usd":4.4,"cached_input_per_mtok_usd":0.275}},{"id":"openai/o1","providers":["openai"],"provider_details":[{"provider":"openai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":39.7,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":15.2,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.0701,"lcr":0.65,"gpqa":0.747474747474747,"tau2":0.625730994152047,"aime25":null,"critpt":0.00285714285714286,"gdpval":null,"suffix":null,"ifbench":0.703401360544218,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-11.05,"analystAgent":null,"frontierCode":null,"cacheHitPrice":7.5,"livecodebench":0.679365079365079,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":15.2309422339403,"terminalbenchHard":0.128787878787879,"price1mInputTokens":15,"contextWindowTokens":200000,"omniscienceAccuracy":0.3453333333333333,"price1mOutputTokens":60,"pricePer1k1mpImages":10.125,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":37.5,"price1mBlended0To3To1":26.25,"price1mBlended7To2To1":14.25,"cacheHitDiscountPercent":0.5,"price1mBlended0To100To1":15.445544554455445,"price1mBlended100To1To1":8.088235294117647,"omniscienceHallucinationRate":0.6962830957230143,"inferenceParametersActiveBillions":null},"name":"o1","slug":"o1","source":"https://artificialanalysis.ai/models/o1","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/o1/benchmarks"},"created":0,"released_at":"2024-12-05","release_source":"https://openai.com/index/introducing-chatgpt-pro/","description":"OpenAI o1 is the first full model in OpenAI's o-series, trained with large-scale reinforcement learning to spend more time thinking before it responds. It is aimed at hard reasoning in math, science, and coding, and accepts image input alongside text.","description_sources":[{"url":"https://openrouter.ai/openai/o1","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/o1","title":"OpenAI model documentation"}],"display_name":"OpenAI o1","context_window":200000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":15,"output_per_mtok_usd":60,"cached_input_per_mtok_usd":7.5}},{"id":"openai/gpt-oss-120b","providers":["groq","fireworks","parasail","cerebras"],"provider_details":[{"provider":"groq","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"parasail","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"cerebras","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":11.3,"artificial_analysis_coding_index":30.4,"artificial_analysis_agentic_index":3.7,"artificial_analysis_economics_index":18.3,"artificial_analysis_engineering_index":15.1,"artificial_analysis_intelligence_index":11.6,"artificial_analysis_strategy_and_ops_index":7.9,"artificial_analysis_finance_and_accounting_index":12.2,"artificial_analysis_healthcare_and_medical_index":10.8}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"modelCount":2,"intelligenceIndex":11.6028431512592},"name":"gpt-oss-120b","slug":"gpt-oss-120b","source":"https://artificialanalysis.ai/models/gpt-oss-120b","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-oss-120b/benchmarks"},"created":0,"released_at":"2025-08-05","release_source":"https://openai.com/index/introducing-gpt-oss/","description":"GPT OSS 120B is OpenAI's open-weight mixture-of-experts language model, released under Apache 2.0 for high-reasoning, agentic, and general-purpose production use. It is the larger of the two gpt-oss models, activating a small fraction of its parameters per token and running on a single 80 GB GPU.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-oss-120b","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-oss-120b","title":"OpenAI model documentation"}],"display_name":"GPT OSS 120B","context_window":128000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.35,"output_per_mtok_usd":0.75}},{"id":"openai/gpt-oss-20b","providers":["groq","parasail","bedrock"],"provider_details":[{"provider":"groq","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"parasail","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"bedrock","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":7.3,"artificial_analysis_coding_index":20.7,"artificial_analysis_agentic_index":1.2,"artificial_analysis_economics_index":11.3,"artificial_analysis_engineering_index":11.7,"artificial_analysis_intelligence_index":9,"artificial_analysis_strategy_and_ops_index":4.8,"artificial_analysis_finance_and_accounting_index":7.3,"artificial_analysis_healthcare_and_medical_index":6.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"modelCount":2,"intelligenceIndex":9.95295342568922},"name":"gpt-oss-20b","slug":"gpt-oss-20b","source":"https://artificialanalysis.ai/models/gpt-oss-20b","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/openai/gpt-oss-20b/benchmarks"},"created":0,"released_at":"2025-08-05","release_source":"https://openai.com/index/introducing-gpt-oss/","description":"GPT OSS 20B is the smaller of OpenAI's open-weight gpt-oss models, released under Apache 2.0 for local, on-device, and low-latency use. It uses a mixture-of-experts design with few active parameters per token, running on consumer hardware where GPT OSS 120B needs a datacenter GPU.","description_sources":[{"url":"https://openrouter.ai/openai/gpt-oss-20b","title":"OpenRouter model listing"},{"url":"https://developers.openai.com/api/docs/models/gpt-oss-20b","title":"OpenAI model documentation"}],"display_name":"GPT OSS 20B","context_window":128000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.075,"output_per_mtok_usd":0.3,"cached_input_per_mtok_usd":0.0375}},{"id":"anthropic/claude-opus-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":57.3,"artificial_analysis_coding_index":78,"artificial_analysis_agentic_index":56.5,"artificial_analysis_economics_index":60.8,"artificial_analysis_engineering_index":54,"artificial_analysis_intelligence_index":50.8,"artificial_analysis_strategy_and_ops_index":56.6,"artificial_analysis_finance_and_accounting_index":54.9,"artificial_analysis_healthcare_and_medical_index":46.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.548656163113994,"lcr":0.793333333333333,"gpqa":0.932323232323232,"tau2":null,"aime25":null,"critpt":0.291428571428571,"gdpval":1720.14,"ifbench":null,"mmmuPro":0.847398843930636,"scicode":0.563657407407407,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.420618556701031,"licenseName":null,"mlcrOverall":0.111111111111111,"omniscience":37.06666666666667,"analystAgent":0.5375,"frontierCode":null,"cacheHitPrice":0.5,"gdpPdfAllPass":0.216,"livecodebench":null,"hostModelCount":8,"cacheWritePrice":6.25,"reasoningTokens":0,"terminalBench21":0.891385767790262,"terminalBench40":0.48989898989899,"enterpriseOpsGym":0.474783646672635,"gdpvalNormalized":0.61007,"intelligenceIndex":50.7771115797629,"terminalbenchHard":null,"briefcaseTotalCost":1619.1916826926013,"price1mInputTokens":5,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.6086666666666667,"price1mOutputTokens":25,"pricePer1k1mpImages":6.86,"terminalBenchScience":0.285714285714286,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":15,"price1mBlended0To3To1":10,"price1mBlended7To2To1":3.85,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":5.198019801980198,"price1mBlended100To1To1":0.7843137254901961,"automationBenchPartialScore":0.565735557649325,"intelligenceIndexTimePerTask":859.7324983859542,"omniscienceHallucinationRate":0.6081771720613288,"inferenceParametersActiveBillions":null},"name":"Claude Opus 5 (Max)","slug":"claude-opus-5","source":"https://artificialanalysis.ai/models/claude-opus-5","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/anthropic/claude-opus-5/benchmarks"},"created":0,"released_at":"2026-07-24","release_source":"https://www.anthropic.com/news/claude-opus-5","description":"Claude Opus 5 is Anthropic's Opus-tier model, frontier intelligence below Claude Fable 5 for everyday professional work in engineering, research, and automation. Compared with Opus 4.8 it verifies its work more carefully, uses fewer tokens, and adds fast mode and automatic safety fallbacks.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-opus-5","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-opus-5","title":"Anthropic release announcement"}],"display_name":"Claude Opus 5","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":true,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":25,"cached_input_per_mtok_usd":0.5,"cache_write_per_mtok_usd":6.25,"cache_write_1h_per_mtok_usd":10}},{"id":"anthropic/claude-opus-5-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":63.3,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":65.6,"artificial_analysis_engineering_index":60.4,"artificial_analysis_intelligence_index":57.6,"artificial_analysis_strategy_and_ops_index":63.7,"artificial_analysis_finance_and_accounting_index":60.7,"artificial_analysis_healthcare_and_medical_index":54.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"modelCount":5,"intelligenceIndex":57.6223698102963},"name":"Claude Opus 5.5","slug":"claude-opus-5-5","source":"https://artificialanalysis.ai/models/claude-opus-5-5","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/anthropic/claude-opus-5-5/benchmarks"},"created":0,"released_at":"2026-09-22","release_source":"https://www.anthropic.com/news/claude-opus-5-5","description":"Claude Opus 5.5 is Anthropic's flagship Opus-tier model and the first release in the Claude 5.5 family, built for long-running agentic coding and knowledge work. It succeeds Claude Opus 5 as the tier's default choice for multi-step changes across large codebases, code review, and financial and scientific analysis.","description_sources":[{"url":"https://www.anthropic.com/news/claude-opus-5-5","title":"Introducing Claude Opus 5.5"},{"url":"https://platform.claude.com/docs/en/about-claude/models/overview","title":"Anthropic models overview"},{"url":"https://openrouter.ai/anthropic/claude-opus-5.5","title":"OpenRouter model listing"}],"display_name":"Claude Opus 5.5","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":false,"default":"medium"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":4,"output_per_mtok_usd":20,"cached_input_per_mtok_usd":0.2,"cache_write_per_mtok_usd":5,"cache_write_1h_per_mtok_usd":8}},{"id":"anthropic/claude-fable-5-1","providers":["anthropic"],"provider_details":[{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":60.5,"artificial_analysis_coding_index":81.6,"artificial_analysis_agentic_index":57.9,"artificial_analysis_economics_index":62.7,"artificial_analysis_engineering_index":56.5,"artificial_analysis_intelligence_index":53.4,"artificial_analysis_strategy_and_ops_index":59.7,"artificial_analysis_finance_and_accounting_index":56.4,"artificial_analysis_healthcare_and_medical_index":49.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"modelCount":5,"intelligenceIndex":53.3549259623252},"name":"Claude Fable 5.1","slug":"claude-fable-5-1","source":"https://artificialanalysis.ai/models/claude-fable-5-1","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/anthropic/claude-fable-5-1/benchmarks"},"created":0,"released_at":"2026-09-01","release_source":"https://www.anthropic.com/claude/fable","description":"Claude Fable 5.1 is Anthropic's flagship Mythos-level model, above the Opus, Sonnet, and Haiku tiers, built for long-running agentic and whole-codebase work. Against Claude Fable 5 it runs faster, completes more tasks, and uses fewer tokens per task.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-fable-5.1","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/claude/fable","title":"Anthropic model page"}],"display_name":"Claude Fable 5.1","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":false,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":10,"output_per_mtok_usd":50,"cached_input_per_mtok_usd":0.25,"cache_write_per_mtok_usd":12.5,"cache_write_1h_per_mtok_usd":20}},{"id":"anthropic/claude-fable-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":58,"artificial_analysis_coding_index":76.5,"artificial_analysis_agentic_index":50.7,"artificial_analysis_economics_index":60.4,"artificial_analysis_engineering_index":53,"artificial_analysis_intelligence_index":49.6,"artificial_analysis_strategy_and_ops_index":55.5,"artificial_analysis_finance_and_accounting_index":53.8,"artificial_analysis_healthcare_and_medical_index":47.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.554680259499537,"lcr":0.823333333333333,"gpqa":0.926262626262626,"tau2":0.985380116959064,"aime25":null,"critpt":0.2857142857,"gdpval":1608.89,"ifbench":0.63469387755102,"mmmuPro":null,"scicode":0.609953703703704,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.381443298969072,"licenseName":null,"mlcrOverall":0.205555555555556,"omniscience":43.3,"analystAgent":0.4875,"frontierCode":null,"cacheHitPrice":1,"gdpPdfAllPass":0.24,"livecodebench":null,"hostModelCount":7,"cacheWritePrice":12.5,"reasoningTokens":0,"terminalBench21":0.846441947565543,"terminalBench40":0.424242424242424,"enterpriseOpsGym":0.5111906893464637,"gdpvalNormalized":0.5544450000000001,"intelligenceIndex":49.6258202733378,"terminalbenchHard":0.628787878787879,"briefcaseTotalCost":2029.603637925242,"price1mInputTokens":10,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.6535,"price1mOutputTokens":50,"pricePer1k1mpImages":13.72,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":30,"price1mBlended0To3To1":20,"price1mBlended7To2To1":7.7,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":10.396039603960396,"price1mBlended100To1To1":1.5686274509803921,"automationBenchPartialScore":0.5407130114065382,"intelligenceIndexTimePerTask":627.0699063634389,"omniscienceHallucinationRate":0.6363636363636364,"inferenceParametersActiveBillions":null},"name":"Claude Fable 5 (Max, Opus 4.8 Fallback)","slug":"claude-fable-5","source":"https://artificialanalysis.ai/models/claude-fable-5","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/anthropic/claude-fable-5/benchmarks"},"created":0,"released_at":"2026-06-09","release_source":"https://www.anthropic.com/news/claude-fable-5-mythos-5","description":"Claude Fable 5 is Anthropic's first generally available Mythos-class model, built for autonomous coding, research, and long-horizon agentic work. It shares its underlying model with Claude Mythos 5 and adds safety measures around dual-use capabilities for general availability.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-fable-5","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-fable-5-mythos-5","title":"Anthropic release announcement"}],"display_name":"Claude Fable 5","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":false,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":10,"output_per_mtok_usd":50,"cached_input_per_mtok_usd":1,"cache_write_per_mtok_usd":12.5,"cache_write_1h_per_mtok_usd":20}},{"id":"anthropic/claude-opus-4-8","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":47.9,"artificial_analysis_coding_index":74.3,"artificial_analysis_agentic_index":41.9,"artificial_analysis_economics_index":53.8,"artificial_analysis_engineering_index":43.5,"artificial_analysis_intelligence_index":41.8,"artificial_analysis_strategy_and_ops_index":44.1,"artificial_analysis_finance_and_accounting_index":46,"artificial_analysis_healthcare_and_medical_index":41.3}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:06:51.922Z"},"data":{"hle":0.486561631139944,"lcr":0.776666666666667,"gpqa":0.92020202020202,"tau2":0.944444444444445,"aime25":null,"critpt":0.208571428571429,"gdpval":1452.8,"ifbench":0.622448979591837,"mmmuPro":null,"scicode":0.543981481481482,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.342268041237113,"licenseName":null,"mlcrOverall":0.194444444444444,"omniscience":28.75,"analystAgent":0.45,"frontierCode":null,"cacheHitPrice":0.5,"gdpPdfAllPass":0.228,"livecodebench":null,"hostModelCount":7,"cacheWritePrice":6.25,"reasoningTokens":0,"terminalBench21":0.846441947565543,"terminalBench40":0.217171717171717,"enterpriseOpsGym":0.4395702775290958,"gdpvalNormalized":0.4764,"intelligenceIndex":41.7899103794639,"terminalbenchHard":0.583333333333333,"briefcaseTotalCost":751.8443173062924,"price1mInputTokens":5,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.48833333333333334,"price1mOutputTokens":25,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":15,"price1mBlended0To3To1":10,"price1mBlended7To2To1":3.85,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":5.198019801980198,"price1mBlended100To1To1":0.7843137254901961,"automationBenchPartialScore":0.45593436915277186,"intelligenceIndexTimePerTask":706.6445506122644,"omniscienceHallucinationRate":0.3925081433224756,"inferenceParametersActiveBillions":null},"name":"Claude Opus 4.8 (Max)","slug":"claude-opus-4-8","source":"https://artificialanalysis.ai/models/claude-opus-4-8","fetchedAt":"2026-10-22T16:06:51.922Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:06:51.922Z","fetchedAt":"2026-10-22T16:06:51.922Z","nextCheckAt":"2026-10-29T16:06:51.922Z","summary":true,"detailPath":"/v1/models/anthropic/claude-opus-4-8/benchmarks"},"created":0,"released_at":"2026-05-28","release_source":"https://www.anthropic.com/news/claude-opus-4-8","description":"Claude Opus 4.8 is an Opus-tier model from Anthropic aimed at agentic coding, legal and financial document work, and complex multi-step reasoning in enterprise workflows. Over Opus 4.7 it is far more willing to flag flaws and uncertainties, catches its own mistakes during agentic runs, and uses tokens more efficiently.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-opus-4.8","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-opus-4-8","title":"Anthropic release announcement"}],"display_name":"Claude Opus 4.8","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":true,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":25,"cached_input_per_mtok_usd":0.5,"cache_write_per_mtok_usd":6.25,"cache_write_1h_per_mtok_usd":10}},{"id":"anthropic/claude-opus-4-7","providers":["anthropic"],"provider_details":[{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":73.6,"artificial_analysis_agentic_index":38.6,"artificial_analysis_economics_index":50,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":40.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.423076923076923,"lcr":0.786666666666667,"gpqa":0.914141414141414,"tau2":0.885964912280702,"aime25":null,"critpt":0.12,"gdpval":1352.39,"ifbench":0.586394557823129,"mmmuPro":0.788439306358382,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.466572504708098,"licenseUrl":null,"parameters":null,"tauBanking":0.34639175257732,"licenseName":null,"omniscience":27.266666666666666,"analystAgent":0.4375,"frontierCode":null,"cacheHitPrice":0.5,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":6.25,"reasoningTokens":0,"terminalBench21":0.831460674157303,"terminalBench40":null,"gdpvalNormalized":0.42619500000000005,"intelligenceIndex":40.6897928205908,"terminalbenchHard":0.515151515151515,"briefcaseTotalCost":1302.702528425893,"price1mInputTokens":5,"contextWindowTokens":1000000,"omniscienceAccuracy":0.48883333333333334,"price1mOutputTokens":25,"pricePer1k1mpImages":6.7,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":15,"price1mBlended0To3To1":10,"price1mBlended7To2To1":3.85,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":5.198019801980198,"price1mBlended100To1To1":0.7843137254901961,"omniscienceHallucinationRate":0.4228888164329964,"inferenceParametersActiveBillions":null},"name":"Claude Opus 4.7 (Max)","slug":"claude-opus-4-7","source":"https://artificialanalysis.ai/models/claude-opus-4-7","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-opus-4-7/benchmarks"},"created":0,"released_at":"2026-04-16","release_source":"https://www.anthropic.com/news/claude-opus-4-7","description":"Claude Opus 4.7 is an Opus-tier model from Anthropic built for demanding software engineering, long-running agents, and vision-heavy tasks like computer use. It improves on Opus 4.6 with higher-resolution image input, an updated tokenizer, a new xhigh effort level, and stricter instruction following.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-opus-4.7","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-opus-4-7","title":"Anthropic release announcement"}],"display_name":"Claude Opus 4.7","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":true,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":25,"cached_input_per_mtok_usd":0.5,"cache_write_per_mtok_usd":6.25,"cache_write_1h_per_mtok_usd":10}},{"id":"anthropic/claude-opus-4-6","providers":["bedrock","anthropic"],"provider_details":[{"provider":"bedrock","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":26.4,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.190917516218721,"lcr":0.67,"gpqa":0.84040404040404,"tau2":0.847953216374269,"aime25":null,"critpt":0.0284081632653061,"gdpval":null,"ifbench":0.446258503401361,"mmmuPro":0.725433526011561,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":2.3666666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.5,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":6.25,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":26.3534088054787,"terminalbenchHard":0.484848484848485,"price1mInputTokens":5,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.4578333333333333,"price1mOutputTokens":25,"pricePer1k1mpImages":6.67,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":15,"price1mBlended0To3To1":10,"price1mBlended7To2To1":3.85,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":5.198019801980198,"price1mBlended100To1To1":0.7843137254901961,"omniscienceHallucinationRate":0.8007992622194897,"inferenceParametersActiveBillions":null},"name":"Claude Opus 4.6 (Non-reasoning, High)","slug":"claude-opus-4-6","source":"https://artificialanalysis.ai/models/claude-opus-4-6","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-opus-4-6/benchmarks"},"created":0,"released_at":"2026-02-05","release_source":"https://www.anthropic.com/news/claude-opus-4-6","description":"Claude Opus 4.6 is an Opus-tier model from Anthropic built for agentic coding, large-codebase navigation, financial and research work, and long-horizon planning. It introduced adaptive thinking, which lets the model decide when to reason at length, along with a much larger context window, context compaction, and parallel agent teams.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-opus-4.6","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-opus-4-6","title":"Anthropic release announcement"}],"display_name":"Claude Opus 4.6","context_window":1000000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":25,"cached_input_per_mtok_usd":0.5,"cache_write_per_mtok_usd":6.25,"cache_write_1h_per_mtok_usd":10}},{"id":"anthropic/claude-sonnet-5-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":56.4,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":60.8,"artificial_analysis_engineering_index":58.4,"artificial_analysis_intelligence_index":56,"artificial_analysis_strategy_and_ops_index":59.6,"artificial_analysis_finance_and_accounting_index":57.2,"artificial_analysis_healthcare_and_medical_index":53.1}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"modelCount":5,"intelligenceIndex":56.0001286580794},"name":"Claude Sonnet 5.5","slug":"claude-sonnet-5-5","source":"https://artificialanalysis.ai/models/claude-sonnet-5-5","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-sonnet-5-5/benchmarks"},"created":0,"released_at":"2026-09-28","release_source":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview","description":"Claude Sonnet 5.5 is Anthropic's Sonnet-tier model and the direct successor to Claude Sonnet 5, sitting between Haiku and Opus as the family's balance of speed and intelligence. It targets well-scoped everyday work such as building features, fixing bugs, and drafting documents and spreadsheets, and it differs from its predecessor in keeping thinking always on, so effort becomes the main lever for trading depth against latency.","description_sources":[{"url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview","title":"Claude Sonnet 5.5 model overview"},{"url":"https://platform.claude.com/docs/en/models/overview","title":"Anthropic models overview"},{"url":"https://openrouter.ai/anthropic/claude-sonnet-5.5","title":"OpenRouter model listing"}],"display_name":"Claude Sonnet 5.5","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":false,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.2,"cache_write_per_mtok_usd":2.5,"cache_write_1h_per_mtok_usd":4}},{"id":"anthropic/claude-sonnet-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":42.3,"artificial_analysis_coding_index":71.5,"artificial_analysis_agentic_index":43.6,"artificial_analysis_economics_index":49.4,"artificial_analysis_engineering_index":40.2,"artificial_analysis_intelligence_index":38.2,"artificial_analysis_strategy_and_ops_index":39.2,"artificial_analysis_finance_and_accounting_index":40.4,"artificial_analysis_healthcare_and_medical_index":35.3}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.412882298424467,"lcr":0.82,"gpqa":0.911111111111111,"tau2":null,"aime25":null,"critpt":0.168571428571429,"gdpval":1462.47,"ifbench":null,"mmmuPro":0.772832369942197,"scicode":0.542824074074074,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.37319587628866,"licenseName":null,"mlcrOverall":0.05,"omniscience":16.45,"analystAgent":0.4625,"frontierCode":null,"cacheHitPrice":0.2,"gdpPdfAllPass":0.132,"livecodebench":null,"hostModelCount":9,"cacheWritePrice":2.5,"reasoningTokens":0,"terminalBench21":0.805243445692884,"terminalBench40":0.141414141414141,"enterpriseOpsGym":0.4467323187108326,"gdpvalNormalized":0.481235,"intelligenceIndex":38.1638712882576,"terminalbenchHard":null,"briefcaseTotalCost":1313.0550111096054,"price1mInputTokens":2,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.4005,"price1mOutputTokens":10,"pricePer1k1mpImages":4.116,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":6,"price1mBlended0To3To1":4,"price1mBlended7To2To1":1.54,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.0792079207920793,"price1mBlended100To1To1":0.3137254901960784,"automationBenchPartialScore":0.36512675717036386,"intelligenceIndexTimePerTask":873.6340980525897,"omniscienceHallucinationRate":0.39366138448707255,"inferenceParametersActiveBillions":null},"name":"Claude Sonnet 5 (Max)","slug":"claude-sonnet-5","source":"https://artificialanalysis.ai/models/claude-sonnet-5","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-sonnet-5/benchmarks"},"created":0,"released_at":"2026-06-30","release_source":"https://www.anthropic.com/news/claude-sonnet-5","description":"Claude Sonnet 5 is Anthropic's mid-tier Sonnet model, positioned for multi-step coding, business automation, legal research, and computer-use agents at lower cost than Opus. Anthropic calls it the most agentic Sonnet yet: it finishes tasks earlier Sonnets stopped short on, verifies its work unprompted, and offers selectable reasoning effort levels.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-sonnet-5","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-sonnet-5","title":"Anthropic release announcement"}],"display_name":"Claude Sonnet 5","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":true,"default":"high"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.2,"cache_write_per_mtok_usd":2.5,"cache_write_1h_per_mtok_usd":4}},{"id":"anthropic/claude-sonnet-4-6","providers":["bedrock","anthropic"],"provider_details":[{"provider":"bedrock","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":24.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.133456904541242,"lcr":0.683333333333333,"gpqa":0.798989898989899,"tau2":0.795321637426901,"aime25":null,"critpt":0.00857142857142857,"gdpval":null,"ifbench":0.41156462585034,"mmmuPro":0.705780346820809,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-3.55,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.3,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":3.75,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":24.6869297210126,"terminalbenchHard":0.462121212121212,"price1mInputTokens":3,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.3855,"price1mOutputTokens":15,"pricePer1k1mpImages":4,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":9,"price1mBlended0To3To1":6,"price1mBlended7To2To1":2.31,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":3.118811881188119,"price1mBlended100To1To1":0.47058823529411764,"omniscienceHallucinationRate":0.6851098454027664,"inferenceParametersActiveBillions":null},"name":"Claude Sonnet 4.6 (Non-reasoning, High)","slug":"claude-sonnet-4-6","source":"https://artificialanalysis.ai/models/claude-sonnet-4-6","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-sonnet-4-6/benchmarks"},"created":0,"released_at":"2026-02-17","release_source":"https://www.anthropic.com/news/claude-sonnet-4-6","description":"Claude Sonnet 4.6 is Anthropic's mid-tier Sonnet model for coding, computer use, document comprehension, and office automation. Compared with Sonnet 4.5 it is less prone to overengineering, follows instructions more closely, hallucinates less, and gains a much larger context window in beta.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-sonnet-4.6","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-sonnet-4-6","title":"Anthropic release announcement"}],"display_name":"Claude Sonnet 4.6","context_window":1000000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":3,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.3,"cache_write_per_mtok_usd":3.75,"cache_write_1h_per_mtok_usd":6}},{"id":"anthropic/claude-haiku-5-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":42.3,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":51.2,"artificial_analysis_engineering_index":44.9,"artificial_analysis_intelligence_index":43.4,"artificial_analysis_strategy_and_ops_index":40.4,"artificial_analysis_finance_and_accounting_index":43.9,"artificial_analysis_healthcare_and_medical_index":39.9}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"modelCount":5,"intelligenceIndex":43.3950199670746},"name":"Claude Haiku 5.5","slug":"claude-haiku-5-5","source":"https://artificialanalysis.ai/models/claude-haiku-5-5","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-haiku-5-5/benchmarks"},"created":0,"released_at":"2026-10-07","release_source":"https://platform.claude.com/docs/en/models/haiku-5-5/overview","description":"Claude Haiku 5.5 is Anthropic's fastest and cheapest Claude, built for high-volume, latency-sensitive work such as classification, routing, extraction and subagent tasks. Unlike Haiku 4.5 it thinks adaptively by default and is steered with the effort parameter, so the same model covers quick lookups and work that needs a moment's reasoning.","description_sources":[{"url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview","title":"Claude Haiku 5.5 model overview"},{"url":"https://platform.claude.com/docs/en/release-notes/api","title":"Claude API release notes"},{"url":"https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude","title":"Claude models on Google Cloud"}],"display_name":"Claude Haiku 5.5","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":true,"default":"medium"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0.5,"cached_input_per_mtok_usd":0.01,"cache_write_per_mtok_usd":0.125,"cache_write_1h_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":100000,"input_per_mtok_usd":0.5,"output_per_mtok_usd":2.5,"cached_input_per_mtok_usd":0.05,"cache_write_per_mtok_usd":0.625,"cache_write_1h_per_mtok_usd":1}]}},{"id":"anthropic/claude-haiku-4-5","providers":["azure-foundry","bedrock","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"bedrock","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":15.4,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.0421686746987952,"lcr":0.496666666666667,"gpqa":0.646464646464646,"tau2":0.324561403508772,"aime25":0.39,"critpt":0,"gdpval":null,"ifbench":0.420408163265306,"mmmuPro":0.551445086705202,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-7.566666666666666,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.1,"livecodebench":0.511111111111111,"hostModelCount":4,"cacheWritePrice":1.25,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":15.410949885381,"terminalbenchHard":0.272727272727273,"price1mInputTokens":1,"contextWindowTokens":200000,"omniscienceAccuracy":0.14433333333333334,"price1mOutputTokens":5,"pricePer1k1mpImages":1.334,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":3,"price1mBlended0To3To1":2,"price1mBlended7To2To1":0.77,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.0396039603960396,"price1mBlended100To1To1":0.1568627450980392,"omniscienceHallucinationRate":0.2571094663030775,"inferenceParametersActiveBillions":null},"name":"Claude 4.5 Haiku (Non-reasoning)","slug":"claude-4-5-haiku","source":"https://artificialanalysis.ai/models/claude-4-5-haiku","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-haiku-4-5/benchmarks"},"created":0,"released_at":"2025-10-15","release_source":"https://www.anthropic.com/news/claude-haiku-4-5","description":"Claude Haiku 4.5 is Anthropic's small, fast model, built for real-time chat assistants, pair programming, and cost-sensitive workloads where responsiveness matters. It brings near-frontier coding and computer-use ability to the Haiku tier, supports extended thinking, and works well as a subagent handling parallel subtasks orchestrated by a larger Claude model.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-haiku-4.5","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-haiku-4-5","title":"Anthropic release announcement"}],"display_name":"Claude Haiku 4.5","context_window":200000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.1,"cache_write_per_mtok_usd":1.25,"cache_write_1h_per_mtok_usd":2}},{"id":"anthropic/claude-opus-4-5","providers":["anthropic"],"provider_details":[{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":23.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.131603336422614,"lcr":0.706666666666667,"gpqa":0.81010101010101,"tau2":0.862573099415205,"aime25":0.626666666666667,"critpt":0.00285714285714286,"gdpval":null,"ifbench":0.429931972789116,"mmmuPro":0.712138728323699,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-4.133333333333334,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.5,"livecodebench":0.737566137566138,"hostModelCount":4,"cacheWritePrice":6.25,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":23.6788975349785,"terminalbenchHard":0.409090909090909,"price1mInputTokens":5,"contextWindowTokens":200000,"omniscienceAccuracy":0.4091666666666667,"price1mOutputTokens":25,"pricePer1k1mpImages":6.67,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":15,"price1mBlended0To3To1":10,"price1mBlended7To2To1":3.85,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":5.198019801980198,"price1mBlended100To1To1":0.7843137254901961,"omniscienceHallucinationRate":0.7624823695345557,"inferenceParametersActiveBillions":null},"name":"Claude Opus 4.5 (Non-reasoning)","slug":"claude-opus-4-5","source":"https://artificialanalysis.ai/models/claude-opus-4-5","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-opus-4-5/benchmarks"},"created":0,"released_at":"2025-11-24","release_source":"https://www.anthropic.com/news/claude-opus-4-5","description":"Claude Opus 4.5 is an Opus-tier model from Anthropic positioned for coding, long-horizon agents, computer use, and deep research. Its notable advance is token efficiency: a new effort parameter lets developers trade speed and cost against capability, and at medium effort it matches Sonnet 4.5 with far fewer output tokens.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-opus-4.5","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-opus-4-5","title":"Anthropic release announcement"}],"display_name":"Claude Opus 4.5","context_window":200000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":5,"output_per_mtok_usd":25,"cached_input_per_mtok_usd":0.5,"cache_write_per_mtok_usd":6.25,"cache_write_1h_per_mtok_usd":10}},{"id":"anthropic/claude-sonnet-4-5","providers":["azure-foundry","anthropic"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"anthropic","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":19.3,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.072289156626506,"lcr":0.54,"gpqa":0.727272727272727,"tau2":0.704678362573099,"aime25":0.37,"critpt":0,"gdpval":null,"ifbench":0.426530612244898,"mmmuPro":0.652023121387283,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-9.366666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.3,"livecodebench":0.59047619047619,"hostModelCount":4,"cacheWritePrice":3.75,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":19.3436534007474,"terminalbenchHard":0.287878787878788,"price1mInputTokens":3,"contextWindowTokens":1000000,"omniscienceAccuracy":0.2838333333333333,"price1mOutputTokens":15,"pricePer1k1mpImages":4.002,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":9,"price1mBlended0To3To1":6,"price1mBlended7To2To1":2.31,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":3.118811881188119,"price1mBlended100To1To1":0.47058823529411764,"omniscienceHallucinationRate":0.5271119385617873,"inferenceParametersActiveBillions":null},"name":"Claude 4.5 Sonnet (Non-reasoning)","slug":"claude-4-5-sonnet","source":"https://artificialanalysis.ai/models/claude-4-5-sonnet","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/anthropic/claude-sonnet-4-5/benchmarks"},"created":0,"released_at":"2025-09-29","release_source":"https://www.anthropic.com/news/claude-sonnet-4-5","description":"Claude Sonnet 4.5 is Anthropic's mid-tier Sonnet model, positioned for real-world coding, building complex agents, and computer use across browsers and spreadsheets. Over Sonnet 4 it sustains focus across much longer multi-step tasks and pairs with a context editing feature and memory tool that let agents run longer.","description_sources":[{"url":"https://openrouter.ai/anthropic/claude-sonnet-4.5","title":"OpenRouter model listing"},{"url":"https://www.anthropic.com/news/claude-sonnet-4-5","title":"Anthropic release announcement"}],"display_name":"Claude Sonnet 4.5","context_window":200000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":3,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.3,"cache_write_per_mtok_usd":3.75,"cache_write_1h_per_mtok_usd":6}},{"id":"google/gemini-4-argon","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":60,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":60.4,"artificial_analysis_engineering_index":55.2,"artificial_analysis_intelligence_index":52.6,"artificial_analysis_strategy_and_ops_index":56.2,"artificial_analysis_finance_and_accounting_index":53.8,"artificial_analysis_healthcare_and_medical_index":47.8}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.570898980537535,"lcr":0.796666666666667,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.271428571428571,"gdpval":1623.11,"aa3dElo":1153.9782,"ifbench":null,"mmmuPro":null,"scicode":0.618055555555556,"harveyLab":0.03888888888888888,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"mlcrOverall":0.138888888888889,"omniscience":42.35,"analystAgent":0.4875,"frontierCode":0.434532,"cacheHitPrice":0.1,"gdpPdfAllPass":0.218,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":0.570707070707071,"gdpvalNormalized":0.5615549999999999,"intelligenceIndex":52.5605655745982,"terminalbenchHard":null,"briefcaseTotalCost":530.3251936207944,"price1mInputTokens":2,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.499,"price1mOutputTokens":10,"pricePer1k1mpImages":null,"terminalBenchScience":0.528571428571429,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":6,"price1mBlended0To3To1":4,"price1mBlended7To2To1":1.47,"cacheHitDiscountPercent":0.95,"price1mBlended0To100To1":2.0792079207920793,"price1mBlended100To1To1":0.21568627450980393,"automationBenchPartialScore":0.7751346945899195,"intelligenceIndexTimePerTask":721.8230903985225,"omniscienceHallucinationRate":0.1506986027944112,"inferenceParametersActiveBillions":null},"name":"Gemini 4 Argon (High)","slug":"gemini-4-argon","source":"https://artificialanalysis.ai/models/gemini-4-argon","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-4-argon/benchmarks"},"created":0,"released_at":"2026-09-30","release_source":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/","description":"Gemini 4 Argon is Google's first Gemini 4 model, a frontier system built to sustain careful reasoning across long, multi-step professional work. It is aimed at real-world software engineering, enterprise knowledge work such as legal and finance, and cybersecurity defense, rather than the quick everyday turns the Flash line handles.","description_sources":[{"url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/","title":"Google announcement: Gemini 4 Argon"},{"url":"https://techcrunch.com/2026/09/30/google-releases-gemini-4-argon-called-its-most-powerful-model-yet/","title":"TechCrunch: Google releases Gemini 4 Argon"},{"url":"https://www.vals.ai/models/google_gemini-4-argon","title":"Vals AI model listing"}],"display_name":"Gemini 4 Argon","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.1}},{"id":"google/gemini-3.8-flash","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":51.6,"artificial_analysis_coding_index":76.3,"artificial_analysis_agentic_index":40.2,"artificial_analysis_economics_index":52.4,"artificial_analysis_engineering_index":42.8,"artificial_analysis_intelligence_index":40.9,"artificial_analysis_strategy_and_ops_index":48.4,"artificial_analysis_finance_and_accounting_index":46.8,"artificial_analysis_healthcare_and_medical_index":40.1}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"modelCount":3,"intelligenceIndex":40.9262321765904},"name":"Gemini 3.8 Flash","slug":"gemini-3-8-flash","source":"https://artificialanalysis.ai/models/gemini-3-8-flash","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-3.8-flash/benchmarks"},"created":0,"released_at":"2026-09-02","release_source":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/","description":"Gemini 3.8 Flash is Google's current Flash-tier workhorse model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows. It keeps the speed and cost of Gemini 3.7 Flash while working harder on difficult tasks through extra reasoning and iterative tool calls, plus a cyber-defense variant for vetted defenders.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.8-flash","title":"OpenRouter model listing"},{"url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/","title":"Google announcement"}],"display_name":"Gemini 3.8 Flash","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.75,"output_per_mtok_usd":3.75,"cached_input_per_mtok_usd":0.075}},{"id":"google/gemini-3.7-flash","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":50.7,"artificial_analysis_coding_index":76.1,"artificial_analysis_agentic_index":35.2,"artificial_analysis_economics_index":50.6,"artificial_analysis_engineering_index":39.7,"artificial_analysis_intelligence_index":39.1,"artificial_analysis_strategy_and_ops_index":48.3,"artificial_analysis_finance_and_accounting_index":44.9,"artificial_analysis_healthcare_and_medical_index":39.5}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.478683966635774,"lcr":0.816666666666667,"gpqa":0.945454545454545,"tau2":null,"aime25":null,"critpt":0.142857142857143,"gdpval":1391.9,"ifbench":null,"mmmuPro":0.854913294797688,"scicode":0.571759259259259,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.327835051546392,"licenseName":null,"mlcrOverall":0.122222222222222,"omniscience":26.483333333333334,"analystAgent":0.6,"frontierCode":null,"cacheHitPrice":0.075,"gdpPdfAllPass":0.236,"livecodebench":null,"hostModelCount":3,"cacheWritePrice":0.0416666667,"reasoningTokens":0,"terminalBench21":0.857677902621723,"terminalBench40":0.136363636363636,"gdpvalNormalized":0.44595000000000007,"intelligenceIndex":39.0594566404896,"terminalbenchHard":null,"price1mInputTokens":0.75,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.5531666666666667,"price1mOutputTokens":3.75,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":2.25,"price1mBlended0To3To1":1.5,"price1mBlended7To2To1":0.5775,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.7797029702970297,"price1mBlended100To1To1":0.11764705882352941,"automationBenchPartialScore":0.6202749292053469,"omniscienceHallucinationRate":0.6452816113390526,"inferenceParametersActiveBillions":null},"name":"Gemini 3.7 Flash (High)","slug":"gemini-3-7-flash","source":"https://artificialanalysis.ai/models/gemini-3-7-flash","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-3.7-flash/benchmarks"},"created":0,"released_at":"2026-08-13","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 3.7 Flash is a Flash-tier multimodal model from Google for coding, agentic workflows, and reliable multi-step execution, now superseded by Gemini 3.8 Flash. Google positioned it as its workhorse for coding and agents at launch, with substantial gains over Gemini 3.6 Flash in software engineering and web development.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.7-flash","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 3.7 Flash","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.75,"output_per_mtok_usd":3.75,"cached_input_per_mtok_usd":0.075}},{"id":"google/gemini-robotics-er-2-preview","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T19:36:54.825Z","nextCheckAt":"2026-10-28T19:36:54.825Z","summary":true,"detailPath":"/v1/models/google/gemini-robotics-er-2-preview/benchmarks"},"created":0,"released_at":"2026-07-30","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini Robotics-ER 2 Preview is Google's embodied-reasoning vision-language model for robotics, taking text, image, video, and audio input and producing text for task planning. It supports spatial reasoning, video moment finding, agentic code execution, multi-step tool orchestration, and multi-robot coordination, adding caching, code execution, and computer use over Robotics-ER 1.6.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/models/gemini-robotics-er-2-preview","title":"Google Gemini API model page"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini Robotics-ER 2 Preview","context_window":131072,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.1}},{"id":"google/gemini-3.6-flash","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":46.1,"artificial_analysis_coding_index":69.2,"artificial_analysis_agentic_index":29,"artificial_analysis_economics_index":46,"artificial_analysis_engineering_index":35.8,"artificial_analysis_intelligence_index":34,"artificial_analysis_strategy_and_ops_index":41.5,"artificial_analysis_finance_and_accounting_index":39.3,"artificial_analysis_healthcare_and_medical_index":34.6}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.408248378127896,"lcr":0.8,"gpqa":0.928282828282828,"tau2":null,"aime25":null,"critpt":0.105714285714286,"gdpval":1283.69,"ifbench":null,"mmmuPro":0.832369942196532,"scicode":0.533564814814815,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.298969072164948,"licenseName":null,"mlcrOverall":0.1,"omniscience":22.133333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.15,"gdpPdfAllPass":0.174,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.775280898876405,"terminalBench40":0.0707070707070707,"gdpvalNormalized":0.39184500000000005,"intelligenceIndex":33.9785928920714,"terminalbenchHard":null,"briefcaseTotalCost":141.8502305413536,"price1mInputTokens":0.75,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.49966666666666665,"price1mOutputTokens":3.75,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":2.25,"price1mBlended0To3To1":1.5,"price1mBlended7To2To1":0.63,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.7797029702970297,"price1mBlended100To1To1":0.19117647058823528,"automationBenchPartialScore":0.5301669491509355,"intelligenceIndexTimePerTask":188.82736193188012,"omniscienceHallucinationRate":0.5562958027981346,"inferenceParametersActiveBillions":null},"name":"Gemini 3.6 Flash (High)","slug":"gemini-3-6-flash","source":"https://artificialanalysis.ai/models/gemini-3-6-flash","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-3.6-flash/benchmarks"},"created":0,"released_at":"2026-07-21","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 3.6 Flash is a Flash-tier model from Google for coding, agentic workflows, and web and app development, balancing speed with multimodal capability. It improved token efficiency and agentic planning over Gemini 3.5 Flash at a lower price point, producing polished outputs with fewer unnecessary edits.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.6-flash","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 3.6 Flash","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.75,"output_per_mtok_usd":3.75,"cached_input_per_mtok_usd":0.075}},{"id":"google/gemini-3.5-flash","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":44.3,"artificial_analysis_coding_index":70.1,"artificial_analysis_agentic_index":26,"artificial_analysis_economics_index":45.6,"artificial_analysis_engineering_index":34.7,"artificial_analysis_intelligence_index":32.6,"artificial_analysis_strategy_and_ops_index":37.4,"artificial_analysis_finance_and_accounting_index":38.5,"artificial_analysis_healthcare_and_medical_index":33.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.42678405931418,"lcr":0.733333333333333,"gpqa":0.922222222222222,"tau2":0.953216374269006,"aime25":null,"critpt":0.131428571428571,"gdpval":1200.77,"ifbench":0.763265306122449,"mmmuPro":0.842774566473988,"scicode":0.539351851851852,"apexAgents":0.470501474926254,"hleDiamond":null,"itBenchSre":0.403483992467043,"licenseUrl":null,"parameters":null,"tauBanking":0.321649484536082,"licenseName":null,"mlcrOverall":0.1,"omniscience":21.183333333333334,"analystAgent":0.45,"frontierCode":null,"cacheHitPrice":0.15,"gdpPdfAllPass":0.198,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.786516853932584,"terminalBench40":0.0656565656565657,"enterpriseOpsGym":0.5007460459564309,"gdpvalNormalized":0.350385,"intelligenceIndex":32.5983283695905,"terminalbenchHard":0.409090909090909,"briefcaseTotalCost":863.539040413711,"price1mInputTokens":1.5,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.514,"price1mOutputTokens":9,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5.25,"price1mBlended0To3To1":3.375,"price1mBlended7To2To1":1.3050000000000002,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.5742574257425743,"price1mBlended100To1To1":0.25,"automationBenchPartialScore":0.4207973565652745,"intelligenceIndexTimePerTask":269.40944631332866,"omniscienceHallucinationRate":0.6217421124828533,"inferenceParametersActiveBillions":null},"name":"Gemini 3.5 Flash (High)","slug":"gemini-3-5-flash","source":"https://artificialanalysis.ai/models/gemini-3-5-flash","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-3.5-flash/benchmarks"},"created":0,"released_at":"2026-05-19","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 3.5 Flash is a Flash-tier multimodal model from Google, launched as its workhorse for agentic and coding tasks and now positioned for high-throughput work. It brought near-Pro coding and reasoning to the Flash tier and is optimized for coding proficiency and parallel agentic execution.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.5-flash","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 3.5 Flash","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":1.5,"output_per_mtok_usd":9,"cached_input_per_mtok_usd":0.15}},{"id":"google/gemini-3.5-flash-lite","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":30.3,"artificial_analysis_coding_index":49.3,"artificial_analysis_agentic_index":14.3,"artificial_analysis_economics_index":31.4,"artificial_analysis_engineering_index":24.4,"artificial_analysis_intelligence_index":22.2,"artificial_analysis_strategy_and_ops_index":22.4,"artificial_analysis_finance_and_accounting_index":23.5,"artificial_analysis_healthcare_and_medical_index":23.6}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.188137164040778,"lcr":0.76,"gpqa":0.838383838383838,"tau2":null,"aime25":null,"critpt":0,"gdpval":982.36,"suffix":null,"ifbench":null,"mmmuPro":0.790173410404624,"scicode":0.413194444444444,"harveyLab":0,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.175257731958763,"licenseName":null,"mlcrOverall":0.105555555555556,"omniscience":5.233333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.03,"gdpPdfAllPass":0.136,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.535580524344569,"terminalBench40":0.0101010101010101,"enterpriseOpsGym":0.423455684870188,"gdpvalNormalized":0.24118,"intelligenceIndex":22.1685424839812,"terminalbenchHard":null,"briefcaseTotalCost":15.287497201090599,"price1mInputTokens":0.3,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.2946666666666667,"price1mOutputTokens":2.5,"pricePer1k1mpImages":null,"terminalBenchScience":0,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1.4,"price1mBlended0To3To1":0.85,"price1mBlended7To2To1":0.331,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.3217821782178218,"price1mBlended100To1To1":0.056862745098039215,"automationBenchPartialScore":0.25011270511840955,"intelligenceIndexTimePerTask":38.19762301402365,"omniscienceHallucinationRate":0.3435727788279773,"inferenceParametersActiveBillions":null},"name":"Gemini 3.5 Flash-Lite","slug":"gemini-3-5-flash-lite","source":"https://artificialanalysis.ai/models/gemini-3-5-flash-lite","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-3.5-flash-lite/benchmarks"},"created":0,"released_at":"2026-07-21","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 3.5 Flash-Lite is Google's fastest and most cost-effective model in the Gemini 3.5 family, built for high-throughput execution and high-volume automation. It is pitched as a low-latency subagent option with upgraded agentic capabilities, suited to focused tasks delegated within larger multi-agent workflows.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.5-flash-lite","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 3.5 Flash-Lite","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.3,"output_per_mtok_usd":2.5,"cached_input_per_mtok_usd":0.03}},{"id":"google/gemini-3.1-pro-preview","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":43.7,"artificial_analysis_coding_index":68.8,"artificial_analysis_agentic_index":8.2,"artificial_analysis_economics_index":44.4,"artificial_analysis_engineering_index":33,"artificial_analysis_intelligence_index":29.7,"artificial_analysis_strategy_and_ops_index":30,"artificial_analysis_finance_and_accounting_index":33.5,"artificial_analysis_healthcare_and_medical_index":29.8}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.470342910101946,"lcr":0.82,"gpqa":0.941414141414141,"tau2":0.956140350877193,"aime25":null,"critpt":0.177142857142857,"gdpval":788.97,"suffix":null,"ifbench":0.771428571428571,"mmmuPro":0.824277456647399,"scicode":0.586805555555556,"harveyLab":0,"apexAgents":0.320058997050148,"hleDiamond":null,"itBenchSre":0.303309120258273,"licenseUrl":null,"parameters":null,"tauBanking":0.214432989690722,"licenseName":null,"mlcrOverall":0.111111111111111,"omniscience":31.883333333333333,"analystAgent":0.4125,"frontierCode":null,"cacheHitPrice":0.2,"gdpPdfAllPass":0.178,"livecodebench":null,"hostModelCount":3,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.737827715355805,"terminalBench40":0.0404040404040404,"enterpriseOpsGym":0.42166517457475383,"gdpvalNormalized":0.144485,"intelligenceIndex":29.7185656132261,"terminalbenchHard":0.537878787878788,"briefcaseTotalCost":106.51806864014081,"price1mInputTokens":2,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.5485,"price1mOutputTokens":12,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":7,"price1mBlended0To3To1":4.5,"price1mBlended7To2To1":1.7399999999999998,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.099009900990099,"price1mBlended100To1To1":0.3333333333333333,"automationBenchPartialScore":0.3542771839809836,"intelligenceIndexTimePerTask":135.61505667830633,"omniscienceHallucinationRate":0.5086747877445552,"inferenceParametersActiveBillions":null},"name":"Gemini 3.1 Pro Preview","slug":"gemini-3-1-pro-preview","source":"https://artificialanalysis.ai/models/gemini-3-1-pro-preview","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-3.1-pro-preview/benchmarks"},"created":0,"released_at":"2026-02-19","release_source":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/","description":"Gemini 3.1 Pro (preview) is Google's Pro-tier frontier reasoning model, offered in preview for complex problem-solving, agentic coding, and data synthesis. Over Gemini 3 Pro it delivers a step forward in core reasoning, more reliable agentic behaviour, and leaner token usage.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.1-pro-preview","title":"OpenRouter model listing"},{"url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/","title":"Google announcement"}],"display_name":"Gemini 3.1 Pro (preview)","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":12,"cached_input_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":4,"output_per_mtok_usd":18,"cached_input_per_mtok_usd":0.4}]}},{"id":"google/gemini-3.1-flash-lite","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["gemini-3-1-flash-lite: AA page HTTP 404"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":22.1,"artificial_analysis_coding_index":34.7,"artificial_analysis_agentic_index":1.6,"artificial_analysis_economics_index":23.4,"artificial_analysis_engineering_index":18.5,"artificial_analysis_intelligence_index":15.6,"artificial_analysis_strategy_and_ops_index":13.7,"artificial_analysis_finance_and_accounting_index":16.7,"artificial_analysis_healthcare_and_medical_index":13.5}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-27T19:36:54.825Z"},"data":{"hle":0.17191844300278,"lcr":0.743333333333333,"gpqa":0.822222222222222,"tau2":0.312865497076023,"aime25":null,"critpt":0.0114285714285714,"gdpval":429.98,"suffix":null,"ifbench":0.772108843537415,"mmmuPro":0.755491329479769,"scicode":0.434027777777778,"apexAgents":0.121681415929204,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.0969072164948454,"healthbench":null,"licenseName":null,"mlcrOverall":0.05,"omniscience":-16.416666666666668,"analystAgent":0.0875,"frontierCode":null,"cacheHitPrice":0.025,"gdpPdfAllPass":0.078,"livecodebench":null,"frontierCode20":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.310861423220974,"terminalBench40":0.00505050505050505,"enterpriseOpsGym":0.2802148612354521,"critptAgenticHard":null,"intelligenceIndex":15.554288713025,"terminalbenchHard":0.242424242424242,"briefcaseTotalCost":3.7077477233649,"price1mInputTokens":0.25,"contextWindowTokens":1000000,"omniscienceAccuracy":0.3626666666666667,"price1mOutputTokens":1.5,"pricePer1k1mpImages":0.28,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.875,"price1mBlended0To3To1":0.5625,"price1mBlended7To2To1":0.21749999999999997,"cacheHitDiscountPercent":0.9,"healthbenchProfessional":null,"price1mBlended0To100To1":0.2623762376237624,"price1mBlended100To1To1":0.041666666666666664,"automationBenchPartialScore":0.06796182033119545,"intelligenceIndexTimePerTask":31.170005153152605,"omniscienceHallucinationRate":0.8266213389121339,"inferenceParametersActiveBillions":null},"name":"Gemini 3.1 Flash-Lite","slug":"gemini-3-1-flash-lite-preview","source":"https://artificialanalysis.ai/models/gemini-3-1-flash-lite-preview","fetchedAt":"2026-10-27T19:36:54.825Z","pageRecords":[]}],"checkedAt":"2026-10-27T19:36:54.825Z","fetchedAt":"2026-09-11T11:21:18.106Z","nextCheckAt":"2026-10-28T19:36:54.825Z","summary":true,"detailPath":"/v1/models/google/gemini-3.1-flash-lite/benchmarks"},"created":0,"released_at":"2026-05-07","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 3.1 Flash-Lite is Google's generally available Flash-Lite model in the Gemini 3 family, optimized for speed and cost on high-volume workloads. It accepts text, image, video, audio, and PDF input and is designed for lightweight agentic tasks where a larger model would be overkill.","description_sources":[{"url":"https://openrouter.ai/google/gemini-3.1-flash-lite","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 3.1 Flash-Lite","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.25,"output_per_mtok_usd":1.5,"audio_input_per_mtok_usd":0.5,"cached_input_per_mtok_usd":0.025}},{"id":"google/gemini-2.5-pro","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":23.7,"artificial_analysis_coding_index":33.3,"artificial_analysis_agentic_index":1.6,"artificial_analysis_economics_index":24.5,"artificial_analysis_engineering_index":18.9,"artificial_analysis_intelligence_index":16.1,"artificial_analysis_strategy_and_ops_index":12.6,"artificial_analysis_finance_and_accounting_index":17.3,"artificial_analysis_healthcare_and_medical_index":15}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.225208526413346,"lcr":0.69,"gpqa":0.844444444444444,"tau2":0.54093567251462,"aime25":0.876666666666667,"critpt":0.0257142857142857,"gdpval":452.91,"suffix":null,"ifbench":0.487074829931973,"mmmuPro":0.749132947976879,"scicode":0.462962962962963,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.0969072164948454,"licenseName":null,"mlcrOverall":0.122222222222222,"omniscience":-16.35,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.125,"gdpPdfAllPass":0.102,"livecodebench":0.801058201058201,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.284644194756554,"terminalBench40":0,"gdpvalNormalized":0,"intelligenceIndex":16.0778962941864,"terminalbenchHard":0.265151515151515,"briefcaseTotalCost":45.28939508085375,"price1mInputTokens":1.25,"contextWindowTokens":1000000,"omniscienceAccuracy":0.3905,"price1mOutputTokens":10,"pricePer1k1mpImages":1.29,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5.625,"price1mBlended0To3To1":3.4375,"price1mBlended7To2To1":1.3375,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":1.3366336633663367,"price1mBlended100To1To1":0.23284313725490197,"automationBenchPartialScore":0.021749464193408737,"intelligenceIndexTimePerTask":87.52943418905811,"omniscienceHallucinationRate":0.9089417555373257,"inferenceParametersActiveBillions":null},"name":"Gemini 2.5 Pro","slug":"gemini-2-5-pro","source":"https://artificialanalysis.ai/models/gemini-2-5-pro","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-2.5-pro/benchmarks"},"created":0,"released_at":"2025-06-17","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 2.5 Pro is Google's Pro-tier model in the Gemini 2.5 family, built for advanced reasoning, coding, mathematics, and scientific tasks. It was Google's first generally available Pro model with built-in thinking, reasoning through a response before answering, and remains the deepest reasoning and coding option in the 2.5 generation.","description_sources":[{"url":"https://openrouter.ai/google/gemini-2.5-pro","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 2.5 Pro","context_window":1000000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.125,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":2.5,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.25}]}},{"id":"google/gemini-2.5-flash","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":9.9,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.0472659870250232,"lcr":0.499,"gpqa":0.682828282828283,"tau2":0.149122807017544,"aime25":0.603333333333333,"critpt":0.0142857142857143,"gdpval":null,"ifbench":0.389795918367347,"mmmuPro":0.654913294797688,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-42.56666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.03,"livecodebench":0.495238095238095,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":9.85128154382851,"terminalbenchHard":0.121212121212121,"price1mInputTokens":0.3,"contextWindowTokens":1000000,"omniscienceAccuracy":0.2613333333333333,"price1mOutputTokens":2.5,"pricePer1k1mpImages":0.1935,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1.4,"price1mBlended0To3To1":0.85,"price1mBlended7To2To1":0.331,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.3217821782178218,"price1mBlended100To1To1":0.056862745098039215,"omniscienceHallucinationRate":0.9300541516245487,"inferenceParametersActiveBillions":null},"name":"Gemini 2.5 Flash (Non-reasoning)","slug":"gemini-2-5-flash","source":"https://artificialanalysis.ai/models/gemini-2-5-flash","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-2.5-flash/benchmarks"},"created":0,"released_at":"2025-06-17","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 2.5 Flash is Google's Flash-tier workhorse in the Gemini 2.5 family, positioned as its price-performance choice for low-latency, high-volume tasks that still need reasoning. It carries the family's built-in thinking capability, so it can reason before responding across coding, mathematics, and scientific work at Flash speed.","description_sources":[{"url":"https://openrouter.ai/google/gemini-2.5-flash","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 2.5 Flash","context_window":1000000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.3,"output_per_mtok_usd":2.5,"audio_input_per_mtok_usd":1,"cached_input_per_mtok_usd":0.03}},{"id":"google/gemini-2.5-flash-lite","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":6.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"hle":0.0366079703429101,"lcr":0.32,"gpqa":0.473737373737374,"tau2":0.190058479532164,"aime25":0.353333333333333,"critpt":0,"gdpval":null,"ifbench":0.314965986394558,"mmmuPro":0.539884393063584,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-58.78333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.01,"livecodebench":0.4,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":6.66853092464521,"terminalbenchHard":0.0227272727272727,"price1mInputTokens":0.1,"contextWindowTokens":1000000,"omniscienceAccuracy":0.15433333333333332,"price1mOutputTokens":0.4,"pricePer1k1mpImages":0.1032,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.25,"price1mBlended0To3To1":0.17500000000000002,"price1mBlended7To2To1":0.067,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":0.10297029702970298,"price1mBlended100To1To1":0.014705882352941176,"omniscienceHallucinationRate":0.87761135199054,"inferenceParametersActiveBillions":null},"name":"Gemini 2.5 Flash-Lite (Non-reasoning)","slug":"gemini-2-5-flash-lite","source":"https://artificialanalysis.ai/models/gemini-2-5-flash-lite","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemini-2.5-flash-lite/benchmarks"},"created":0,"released_at":"2025-07-22","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 2.5 Flash-Lite is the lightweight, fastest, and most budget-friendly multimodal model in Google's Gemini 2.5 family, built for ultra-low-latency, cost-sensitive workloads. It trades some capability for throughput, with faster token generation than Gemini 2.5 Flash while keeping the family's optional reasoning mode.","description_sources":[{"url":"https://openrouter.ai/google/gemini-2.5-flash-lite","title":"OpenRouter model listing"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 2.5 Flash-Lite","context_window":1000000,"reasoning":{"kind":"budget","levels":["minimal","low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image","audio","video"],"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0.4,"audio_input_per_mtok_usd":0.3,"cached_input_per_mtok_usd":0.01}},{"id":"google/gemma-4-31b","providers":["parasail"],"provider_details":[{"provider":"parasail","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":13.9,"artificial_analysis_coding_index":43.4,"artificial_analysis_agentic_index":4.2,"artificial_analysis_economics_index":20.6,"artificial_analysis_engineering_index":16.7,"artificial_analysis_intelligence_index":14.7,"artificial_analysis_strategy_and_ops_index":9.3,"artificial_analysis_finance_and_accounting_index":13.6,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"modelCount":2,"intelligenceIndex":14.6689500745828},"name":"Gemma 4 31B","slug":"gemma-4-31b","source":"https://artificialanalysis.ai/models/gemma-4-31b","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/google/gemma-4-31b/benchmarks"},"created":0,"released_at":"2026-04-02","release_source":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/","description":"Gemma 4 31B is the largest dense open-weight model in Google's Gemma 4 family, released under Apache 2.0 for advanced reasoning and agentic workflows. Unlike the smaller E2B and E4B variants it takes text and image but not audio, favouring quality over speed.","description_sources":[{"url":"https://ai.google.dev/gemma/docs/core","title":"Google announcement"},{"url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/","title":"Google announcement"}],"display_name":"Gemma 4 31B","context_window":128000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.99,"output_per_mtok_usd":1.49}},{"id":"google/gemini-3.8-flash-tts","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/text-to-speech/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-24T17:11:52.155Z"},"data":{"elo":1280,"ci_95":14},"name":"Gemini 3.8 Flash TTS (text-to-speech)","slug":"text-to-speech:gemini-3-8-flash-tts","source":"https://artificialanalysis.ai/api/v2/media/text-to-speech/models/free","fetchedAt":"2026-10-24T17:11:52.155Z","pageRecords":[]}],"checkedAt":"2026-10-24T17:11:52.155Z","fetchedAt":"2026-10-24T17:11:52.155Z","nextCheckAt":"2026-10-31T17:11:52.155Z","summary":true,"detailPath":"/v1/models/google/gemini-3.8-flash-tts/benchmarks"},"created":0,"description":"Gemini 3.8 Flash TTS is Google's text-to-speech model in the Gemini 3.8 family, turning text prompts into expressive 24 kHz speech across 30 prebuilt voices. It is steerable in natural language, so style, pace, accent and tone follow the prompt rather than fixed presets.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/speech-generation","title":"Google Gemini API text-to-speech guide"},{"url":"https://ai.google.dev/gemini-api/docs/models","title":"Google Gemini API model list"}],"display_name":"Gemini 3.8 Flash TTS","output_modality":"audio_speech","input_modalities":["text"],"output_modalities":["audio"],"endpoints":["/v1/audio/speech"],"voices":["Zephyr","Puck","Charon","Kore","Fenrir","Leda","Orus","Aoede","Callirrhoe","Autonoe","Enceladus","Iapetus","Umbriel","Algieba","Despina","Erinome","Algenib","Rasalgethi","Laomedeia","Achernar","Alnilam","Schedar","Gacrux","Pulcherrima","Achird","Zubenelgenubi","Vindemiatrix","Sadachbia","Sadaltager","Sulafat"],"pricing":{"input_per_mtok_usd":0.5,"output_per_mtok_usd":9}},{"id":"google/gemini-3.1-flash-tts-preview","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T19:36:54.825Z","nextCheckAt":"2026-10-28T19:36:54.825Z","summary":true,"detailPath":"/v1/models/google/gemini-3.1-flash-tts-preview/benchmarks"},"created":0,"released_at":"2026-04-15","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Gemini 3.1 Flash TTS (preview) is Google's text-to-speech model in the Gemini 3.1 family, generating natural, low-latency speech from text prompts. It adds expressive audio tags for precise narration control alongside steerable prompts and improved naturalness, controllability, and multilingual support over earlier Gemini TTS models.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-tts-preview","title":"Google Gemini API model page"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Google Gemini API changelog"}],"display_name":"Gemini 3.1 Flash TTS (preview)","output_modality":"audio_speech","input_modalities":["text"],"output_modalities":["audio"],"endpoints":["/v1/audio/speech"],"voices":["Zephyr","Puck","Charon","Kore","Fenrir","Leda","Orus","Aoede","Callirrhoe","Autonoe","Enceladus","Iapetus","Umbriel","Algieba","Despina","Erinome","Algenib","Rasalgethi","Laomedeia","Achernar","Alnilam","Schedar","Gacrux","Pulcherrima","Achird","Zubenelgenubi","Vindemiatrix","Sadachbia","Sadaltager","Sulafat"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":20}},{"id":"xai/grok-4.7","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":54.1,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":55.7,"artificial_analysis_engineering_index":47.4,"artificial_analysis_intelligence_index":46.4,"artificial_analysis_strategy_and_ops_index":54.4,"artificial_analysis_finance_and_accounting_index":51.9,"artificial_analysis_healthcare_and_medical_index":47.9}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T16:37:53.320Z"},"data":{"modelCount":3,"intelligenceIndex":46.4465506302286},"name":"Grok 4.7","slug":"grok-4-7","source":"https://artificialanalysis.ai/models/grok-4-7","fetchedAt":"2026-10-22T16:37:53.320Z","pageRecords":[]}],"checkedAt":"2026-10-22T16:37:53.320Z","fetchedAt":"2026-10-22T16:37:53.320Z","nextCheckAt":"2026-10-29T16:37:53.320Z","summary":true,"detailPath":"/v1/models/xai/grok-4.7/benchmarks"},"created":0,"released_at":"2026-09-21","release_source":"https://x.ai/news/grok-4-7","description":"Grok 4.7 is SpaceXAI's flagship model for coding, agentic tool calling, and knowledge work, accepting text and image input and returning text. It sits above Grok 4.6 on a larger base model trained for longer-running tasks, and adds an xhigh reasoning rung its predecessors do not expose.","description_sources":[{"url":"https://x.ai/news/grok-4-7","title":"Introducing Grok 4.7"},{"url":"https://docs.x.ai/docs/models","title":"xAI model documentation"},{"url":"https://vercel.com/ai-gateway/models/grok-4.7","title":"Vercel AI Gateway model listing"}],"display_name":"Grok 4.7","context_window":500000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":6,"cached_input_per_mtok_usd":0.5,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":4,"output_per_mtok_usd":12,"cached_input_per_mtok_usd":1}]}},{"id":"xai/grok-4.6","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":51.9,"artificial_analysis_coding_index":76.8,"artificial_analysis_agentic_index":53,"artificial_analysis_economics_index":53.9,"artificial_analysis_engineering_index":45,"artificial_analysis_intelligence_index":44.3,"artificial_analysis_strategy_and_ops_index":52.5,"artificial_analysis_finance_and_accounting_index":49.3,"artificial_analysis_healthcare_and_medical_index":45.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.429101019462465,"lcr":0.803333333333333,"gpqa":0.94949494949495,"tau2":null,"aime25":null,"critpt":0.171428571428571,"gdpval":1618.44,"ifbench":null,"mmmuPro":null,"scicode":0.564814814814815,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.507216494845361,"healthbench":null,"licenseName":null,"mlcrOverall":0.155555555555556,"omniscience":30.483333333333334,"analystAgent":0.4125,"frontierCode":null,"cacheHitPrice":0.5,"gdpPdfAllPass":0.17,"livecodebench":null,"frontierCode20":null,"hostModelCount":4,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.883895131086142,"terminalBench40":0.212121212121212,"enterpriseOpsGym":0.4834377797672337,"critptAgenticHard":null,"intelligenceIndex":44.3113073012592,"terminalbenchHard":null,"briefcaseTotalCost":360.237118846159,"price1mInputTokens":2,"contextWindowTokens":500000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.48233333333333334,"price1mOutputTokens":6,"pricePer1k1mpImages":null,"terminalBenchScience":0.0619047619047619,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.35,"cacheHitDiscountPercent":0.75,"healthbenchProfessional":null,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.5686274509803921,"automationBenchPartialScore":0.6667849638121398,"intelligenceIndexTimePerTask":469.1521034081896,"omniscienceHallucinationRate":0.3428847392144237,"inferenceParametersActiveBillions":null},"name":"Grok 4.6 (High)","slug":"grok-4-6","source":"https://artificialanalysis.ai/models/grok-4-6","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/xai/grok-4.6/benchmarks"},"created":0,"released_at":"2026-08-12","release_source":"https://docs.x.ai/developers/release-notes","description":"Grok 4.6 is xAI's frontier flagship Grok model, positioned for coding, agentic tasks, knowledge work, and STEM. It accepts text and image inputs and supports tool calling, with reasoning effort tunable from low through an extra-high xhigh setting that Grok 4.5 does not offer.","description_sources":[{"url":"https://openrouter.ai/x-ai/grok-4.6","title":"OpenRouter model listing"},{"url":"https://docs.x.ai/developers/release-notes","title":"xAI release notes"}],"display_name":"Grok 4.6","context_window":500000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":6,"cached_input_per_mtok_usd":0.5,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":4,"output_per_mtok_usd":12,"cached_input_per_mtok_usd":1}]}},{"id":"xai/grok-4.5","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":48.2,"artificial_analysis_coding_index":72.4,"artificial_analysis_agentic_index":41.2,"artificial_analysis_economics_index":50,"artificial_analysis_engineering_index":39.3,"artificial_analysis_intelligence_index":38.8,"artificial_analysis_strategy_and_ops_index":47,"artificial_analysis_finance_and_accounting_index":44,"artificial_analysis_healthcare_and_medical_index":40.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.42678405931418,"lcr":0.793333333333333,"gpqa":0.931313131313131,"tau2":null,"aime25":null,"critpt":0.154285714285714,"gdpval":1380.98,"ifbench":null,"mmmuPro":0.804046242774566,"scicode":0.549768518518518,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.420618556701031,"healthbench":null,"licenseName":null,"mlcrOverall":0.161111111111111,"omniscience":25.316666666666666,"analystAgent":0.35,"frontierCode":null,"cacheHitPrice":0.3,"gdpPdfAllPass":0.188,"livecodebench":null,"frontierCode20":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.816479400749064,"terminalBench40":0.106060606060606,"enterpriseOpsGym":0.40823634735899733,"critptAgenticHard":null,"intelligenceIndex":38.8121153240657,"terminalbenchHard":null,"briefcaseTotalCost":87.10653419351267,"price1mInputTokens":2,"contextWindowTokens":500000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.5155,"price1mOutputTokens":6,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.21,"cacheHitDiscountPercent":0.85,"healthbenchProfessional":null,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.37254901960784315,"automationBenchPartialScore":0.5793486035084273,"intelligenceIndexTimePerTask":392.34566899291946,"omniscienceHallucinationRate":0.5414516683866529,"inferenceParametersActiveBillions":null},"name":"Grok 4.5 (High)","slug":"grok-4-5","source":"https://artificialanalysis.ai/models/grok-4-5","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/xai/grok-4.5/benchmarks"},"created":0,"released_at":"2026-07-08","release_source":"https://docs.x.ai/developers/release-notes","description":"Grok 4.5 is xAI's Grok 4-generation model for coding, agentic tasks, and knowledge work, sitting below Grok 4.6 in the family. It supports configurable reasoning effort at low, medium, or high, along with tool calling, and xAI serves it from EU-hosted regions.","description_sources":[{"url":"https://openrouter.ai/x-ai/grok-4.5","title":"OpenRouter model listing"},{"url":"https://docs.x.ai/developers/release-notes","title":"xAI release notes"}],"display_name":"Grok 4.5","context_window":500000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":6,"cached_input_per_mtok_usd":0.3,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":4,"output_per_mtok_usd":12,"cached_input_per_mtok_usd":0.6}]}},{"id":"xai/grok-4.3","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":32.3,"artificial_analysis_coding_index":42.2,"artificial_analysis_agentic_index":15.5,"artificial_analysis_economics_index":40.7,"artificial_analysis_engineering_index":30.7,"artificial_analysis_intelligence_index":24.9,"artificial_analysis_strategy_and_ops_index":16.8,"artificial_analysis_finance_and_accounting_index":27,"artificial_analysis_healthcare_and_medical_index":26.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.372103799814643,"lcr":0.73,"gpqa":0.901010101010101,"tau2":0.976608187134503,"aime25":null,"critpt":0.08,"gdpval":937.76,"ifbench":0.812925170068027,"mmmuPro":0.78092485549133,"scicode":0.482638888888889,"apexAgents":0.170353982300885,"hleDiamond":null,"itBenchSre":0.327212806026365,"licenseUrl":null,"parameters":null,"tauBanking":0.123711340206186,"healthbench":null,"licenseName":null,"mlcrOverall":0.0833333333333333,"omniscience":17.983333333333334,"analystAgent":0.0875,"frontierCode":null,"cacheHitPrice":0.2,"gdpPdfAllPass":0.058,"livecodebench":null,"frontierCode20":null,"hostModelCount":3,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.397003745318352,"terminalBench40":0,"enterpriseOpsGym":0.26260817666368247,"critptAgenticHard":null,"intelligenceIndex":24.8801191398002,"terminalbenchHard":0.378787878787879,"briefcaseTotalCost":27.12559001705592,"price1mInputTokens":1.25,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.3478333333333333,"price1mOutputTokens":2.5,"pricePer1k1mpImages":1.28,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1.875,"price1mBlended0To3To1":1.5625,"price1mBlended7To2To1":0.64,"cacheHitDiscountPercent":0.84,"healthbenchProfessional":null,"price1mBlended0To100To1":1.2623762376237624,"price1mBlended100To1To1":0.23284313725490197,"automationBenchPartialScore":0.01375964614032744,"intelligenceIndexTimePerTask":231.70515681230947,"omniscienceHallucinationRate":0.25760286225402507,"inferenceParametersActiveBillions":null},"name":"Grok 4.3 (High)","slug":"grok-4-3","source":"https://artificialanalysis.ai/models/grok-4-3","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/xai/grok-4.3/benchmarks"},"created":0,"description":"Grok 4.3 is xAI's fast, reliable Grok model built for agentic workflows and instruction-following tasks. It takes text and image inputs, supports function calling and structured outputs, and offers selectable reasoning effort from none to xhigh, defaulting to low so responses stay quick.","description_sources":[{"url":"https://openrouter.ai/x-ai/grok-4.3","title":"OpenRouter model listing"},{"url":"https://docs.x.ai/docs/models/grok-4.3","title":"xAI model documentation"}],"display_name":"Grok 4.3","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":2.5,"cached_input_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":2.5,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.4}]}},{"id":"xai/grok-4.20-0309-reasoning","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429","Previous AA match missing from current index; retaining last snapshot."],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":25.2,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.323911028730306,"lcr":0.676666666666667,"gpqa":0.884848484848485,"tau2":0.964912280701754,"aime25":null,"critpt":0.06,"gdpval":null,"ifbench":0.829251700680272,"mmmuPro":0.731791907514451,"scicode":null,"apexAgents":0.14233038348082597,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":12.95,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":25.2162736582441,"terminalbenchHard":0.409090909090909,"price1mInputTokens":2,"contextWindowTokens":2000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.28883333333333333,"price1mOutputTokens":6,"pricePer1k1mpImages":0.514,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.1400000000000001,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.27450980392156865,"omniscienceHallucinationRate":0.22404499648464965,"inferenceParametersActiveBillions":null},"name":"Grok 4.20 0309 (Reasoning)","slug":"grok-4-20-0309","source":"https://artificialanalysis.ai/models/grok-4-20-0309","fetchedAt":"2026-10-14T16:24:48.193Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/xai/grok-4.20-0309-reasoning/benchmarks"},"created":0,"released_at":"2026-03-10","release_source":"https://docs.x.ai/developers/release-notes","description":"Grok 4.20 (reasoning) is the reasoning-enabled snapshot of xAI's Grok 4.20, a general-purpose model aimed at fast agentic tool calling with strict prompt adherence. It accepts text and images, supports function calling and structured outputs, and adds extended reasoning that the non-reasoning snapshot omits.","description_sources":[{"url":"https://docs.x.ai/docs/models/grok-4.20-0309-reasoning","title":"xAI model documentation"},{"url":"https://docs.x.ai/developers/release-notes","title":"xAI release notes"}],"display_name":"Grok 4.20 (reasoning)","context_window":1000000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":2.5,"cached_input_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":2.5,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.4}]}},{"id":"xai/grok-4.20-0309-non-reasoning","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429","Previous AA match missing from current index; retaining last snapshot."],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":14.6,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.245134383688601,"lcr":0.253333333333333,"gpqa":0.784848484848485,"tau2":0.695906432748538,"aime25":null,"critpt":0,"gdpval":null,"ifbench":0.478231292517007,"mmmuPro":0.640462427745665,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-42.583333333333336,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":14.606641173969,"terminalbenchHard":0.21969696969697,"price1mInputTokens":2,"contextWindowTokens":2000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.2605,"price1mOutputTokens":6,"pricePer1k1mpImages":0.514,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.1400000000000001,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.27450980392156865,"omniscienceHallucinationRate":0.9281045751633987,"inferenceParametersActiveBillions":null},"name":"Grok 4.20 0309 (Non-reasoning)","slug":"grok-4-20-0309-non-reasoning","source":"https://artificialanalysis.ai/models/grok-4-20-0309-non-reasoning","fetchedAt":"2026-10-14T16:24:48.193Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/xai/grok-4.20-0309-non-reasoning/benchmarks"},"created":0,"released_at":"2026-03-10","release_source":"https://docs.x.ai/developers/release-notes","description":"Grok 4.20 (non-reasoning) is the non-reasoning snapshot of xAI's Grok 4.20, built for fast, precise responses with agentic tool calling and strict prompt adherence. It shares the reasoning variant's text and image inputs, function calling, and structured outputs, but skips extended reasoning for lower latency.","description_sources":[{"url":"https://docs.x.ai/docs/models/grok-4.20-0309-non-reasoning","title":"xAI model documentation"},{"url":"https://docs.x.ai/developers/release-notes","title":"xAI release notes"}],"display_name":"Grok 4.20 (non-reasoning)","context_window":1000000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":2.5,"cached_input_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":2.5,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.4}]}},{"id":"xai/grok-4.20-multi-agent-0309","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:07:54.046Z","nextCheckAt":"2026-10-28T20:07:54.046Z","summary":true,"detailPath":"/v1/models/xai/grok-4.20-multi-agent-0309/benchmarks"},"created":0,"released_at":"2026-03-10","release_source":"https://docs.x.ai/developers/release-notes","description":"Grok 4.20 (multi-agent) is the multi-agent beta snapshot of xAI's Grok 4.20, aimed at deep research tasks. Instead of a single model pass, multiple agents collaborate in parallel on each request, with the same text and image inputs, function calling, structured outputs, and reasoning as its siblings.","description_sources":[{"url":"https://docs.x.ai/docs/models/grok-4.20-multi-agent-0309","title":"xAI model documentation"},{"url":"https://docs.x.ai/developers/release-notes","title":"xAI release notes"}],"display_name":"Grok 4.20 (multi-agent)","context_window":1000000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":2.5,"cached_input_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":2.5,"output_per_mtok_usd":5,"cached_input_per_mtok_usd":0.4}]}},{"id":"xai/grok-build-0.1","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:07:54.046Z","nextCheckAt":"2026-10-28T20:07:54.046Z","summary":true,"detailPath":"/v1/models/xai/grok-build-0.1/benchmarks"},"created":0,"released_at":"2026-05-19","release_source":"https://docs.x.ai/developers/release-notes","description":"Grok Build 0.1 is xAI's fast coding model, trained specifically for agentic software engineering workflows and released as an early-access preview. It accepts text and image inputs, supports tool calling and reasoning, and is optimized for interactive coding sessions rather than general knowledge work.","description_sources":[{"url":"https://openrouter.ai/x-ai/grok-build-0.1","title":"OpenRouter model listing"},{"url":"https://docs.x.ai/developers/release-notes","title":"xAI release notes"}],"display_name":"Grok Build 0.1","context_window":256000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":2,"cached_input_per_mtok_usd":0.2,"context_price_tiers":[{"above_input_tokens":200000,"input_per_mtok_usd":2,"output_per_mtok_usd":4,"cached_input_per_mtok_usd":0.4}]}},{"id":"xai/grok-voice-tts-1.0","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:07:54.046Z","nextCheckAt":"2026-10-28T20:07:54.046Z","summary":true,"detailPath":"/v1/models/xai/grok-voice-tts-1.0/benchmarks"},"created":0,"description":"Grok Voice TTS 1.0 is xAI's text-to-speech model, turning text into natural, expressive spoken audio for voice agents, accessibility tools, podcasts, and interactive audio. It offers a roster of voices, inline speech tags for laughter, whispers, and pauses, custom voices cloned from a short reference clip, and WebSocket streaming.","description_sources":[{"url":"https://docs.x.ai/developers/model-capabilities/audio/text-to-speech","title":"xAI text-to-speech documentation"},{"url":"https://x.ai/news/grok-stt-and-tts-apis","title":"xAI release announcement"}],"display_name":"Grok Voice TTS 1.0","output_modality":"audio_speech","input_modalities":["text"],"output_modalities":["audio"],"endpoints":["/v1/audio/speech"],"voices":["altair","ara","atlas","carina","castor","celeste","cosmo","eve","helios","helix","iris","kepler","leo","lumen","luna","lux","naksh","orion","perseus","rex","rigel","sal","sirius","ursa","zagan","zenith"],"pricing":{"input_per_mchars_usd":15}},{"id":"meta/llama-3.3-70b","providers":["azure-foundry","parasail"],"provider_details":[{"provider":"azure-foundry","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"parasail","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":11.9,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":7.7,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.0356,"lcr":0.156666666666667,"gpqa":0.497979797979798,"tau2":0.266081871345029,"aime25":0.0766666666666667,"critpt":0,"gdpval":-174.39,"suffix":null,"ifbench":0.470748299319728,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.00564971751412429,"parameters":70,"tauBanking":null,"healthbench":null,"omniscience":-54.166666666666664,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.6,"livecodebench":0.287830687830688,"frontierCode20":null,"hostModelCount":9,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.048689138576779,"terminalBench40":null,"critptAgenticHard":null,"intelligenceIndex":7.66380859299009,"terminalbenchHard":0.0303030303030303,"price1mInputTokens":0.6,"contextWindowTokens":128000,"omniscienceAccuracy":0.1895,"price1mOutputTokens":0.71,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.655,"price1mBlended0To3To1":0.6275,"price1mBlended7To2To1":0.611,"cacheHitDiscountPercent":0.5,"healthbenchProfessional":null,"price1mBlended0To100To1":0.6010891089108911,"price1mBlended100To1To1":0.601078431372549,"omniscienceHallucinationRate":0.9021180341353074,"inferenceParametersActiveBillions":null},"name":"Llama 3.3 Instruct 70B","slug":"llama-3-3-instruct-70b","source":"https://artificialanalysis.ai/models/llama-3-3-instruct-70b","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/meta/llama-3.3-70b/benchmarks"},"created":0,"released_at":"2024-12-06","release_source":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct","description":"Llama 3.3 70B Instruct is Meta's open-weight, text-only instruction-tuned model for multilingual assistant-style chat, commercial and research use, and synthetic data generation. It improves on Llama 3.1 70B, approaches Llama 3.1 405B on some tasks, supports tool calling, and ships only as an instruction-tuned checkpoint.","description_sources":[{"url":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct","title":"Meta Hugging Face model card"},{"url":"https://developer.meta.com/ai/docs/model-cards-and-prompt-formats/llama3_3/","title":"Meta Llama model card"}],"display_name":"Llama 3.3 70B Instruct","context_window":128000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.71,"output_per_mtok_usd":0.71}},{"id":"meta/muse-glimmer-30b","providers":["together"],"provider_details":[{"provider":"together","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:07:54.046Z","nextCheckAt":"2026-10-28T20:07:54.046Z","summary":true,"detailPath":"/v1/models/meta/muse-glimmer-30b/benchmarks"},"created":0,"released_at":"2026-08-10","release_source":"https://research.meta.ai/blog/introducing-muse-glimmer-open-agentic-model","description":"Muse Glimmer 30B is Meta Superintelligence Labs' open-weight, dense multimodal model, distilled from Muse Spark and built for always-on local agents on a single consumer GPU. It handles long-horizon tasks, precise tool calling, interleaved text and image input, failure recovery, and controllable effort, with speculative decoding for faster output.","description_sources":[{"url":"https://openrouter.ai/meta/muse-glimmer-30b","title":"OpenRouter model listing"},{"url":"https://research.meta.ai/blog/introducing-muse-glimmer-open-agentic-model","title":"Meta research announcement"}],"display_name":"Muse Glimmer 30B","context_window":131072,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.35,"output_per_mtok_usd":1.5,"cached_input_per_mtok_usd":0.04}},{"id":"mistral/mistral-large-3","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":11,"artificial_analysis_coding_index":20.1,"artificial_analysis_agentic_index":1,"artificial_analysis_economics_index":13,"artificial_analysis_engineering_index":11.9,"artificial_analysis_intelligence_index":9.3,"artificial_analysis_strategy_and_ops_index":7.5,"artificial_analysis_finance_and_accounting_index":8.9,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.0417052826691381,"lcr":0.36,"gpqa":0.67979797979798,"tau2":0.245614035087719,"aime25":0.38,"critpt":0,"gdpval":419.57,"suffix":null,"ifbench":0.361904761904762,"mmmuPro":0.556647398843931,"scicode":0.365740740740741,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":675,"tauBanking":0.0577319587628866,"healthbench":null,"mlcrOverall":0.0111111111111111,"omniscience":-39.56666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.05,"gdpPdfAllPass":0.012,"livecodebench":0.464550264550264,"frontierCode20":null,"hostModelCount":3,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":0.119850187265918,"terminalBench40":0,"critptAgenticHard":null,"intelligenceIndex":9.26608812860467,"terminalbenchHard":0.159090909090909,"briefcaseTotalCost":4.654419588203055,"price1mInputTokens":0.5,"contextWindowTokens":256000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.2495,"price1mOutputTokens":1.5,"pricePer1k1mpImages":0.668,"terminalBenchScience":null,"price1mBlended0To1To1":1,"price1mBlended0To3To1":0.75,"price1mBlended7To2To1":0.28500000000000003,"cacheHitDiscountPercent":0.9,"healthbenchProfessional":null,"price1mBlended0To100To1":0.5099009900990099,"price1mBlended100To1To1":0.06862745098039216,"automationBenchPartialScore":0.014974701708193462,"intelligenceIndexTimePerTask":65.399445761451,"omniscienceHallucinationRate":0.8596491228070176,"inferenceParametersActiveBillions":41},"name":"Mistral Large 3","slug":"mistral-large-3","source":"https://artificialanalysis.ai/models/mistral-large-3","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/mistral/mistral-large-3/benchmarks"},"created":0,"released_at":"2025-12-02","release_source":"https://legal.mistral.ai/ai-governance/models/mistral-large-3/","description":"Mistral Large 3 is Mistral's open-weight, general-purpose multimodal model, released under Apache 2.0 for chat, document question-answering, and agentic workloads. It uses a granular mixture-of-experts design and supports function calling, structured outputs, built-in agent tools, and batch processing.","description_sources":[{"url":"https://docs.mistral.ai/models/mistral-large-3-25-12","title":"Mistral model documentation"},{"url":"https://legal.mistral.ai/ai-governance/models/mistral-large-3/","title":"Mistral model governance page"}],"display_name":"Mistral Large 3","context_window":128000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.5,"output_per_mtok_usd":1.5}},{"id":"microsoft/phi-4","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":5.9,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.0376,"lcr":0,"gpqa":0.574747474747475,"tau2":0,"aime25":0.18,"critpt":0,"gdpval":null,"suffix":null,"ifbench":0.235374149659864,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":14,"tauBanking":null,"healthbench":null,"omniscience":-55.7,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":0.230687830687831,"frontierCode20":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"critptAgenticHard":null,"intelligenceIndex":5.92257028113751,"terminalbenchHard":0.0378787878787879,"price1mInputTokens":0.125,"contextWindowTokens":16000,"omniscienceAccuracy":0.14066666666666666,"price1mOutputTokens":0.5,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.3125,"price1mBlended0To3To1":0.21875,"price1mBlended7To2To1":0.1625,"cacheHitDiscountPercent":null,"healthbenchProfessional":null,"price1mBlended0To100To1":0.12871287128712872,"price1mBlended100To1To1":0.12867647058823528,"omniscienceHallucinationRate":0.8118696664080682,"inferenceParametersActiveBillions":null},"name":"Phi-4","slug":"phi-4","source":"https://artificialanalysis.ai/models/phi-4","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/microsoft/phi-4/benchmarks"},"created":0,"released_at":"2024-12-12","release_source":"https://techcommunity.microsoft.com/blog/aiplatformblog/introducing-phi-4-microsoft%E2%80%99s-newest-small-language-model-specializing-in-comple/4357090","description":"Phi-4 is Microsoft Research's small open-weight language model, a dense 14-billion-parameter transformer designed for complex reasoning in memory- and latency-constrained settings. It is trained on curated synthetic, textbook-like data for math, coding, and common-sense reasoning, extending the Phi-3 recipe with heavier supervised fine-tuning and preference alignment.","description_sources":[{"url":"https://openrouter.ai/microsoft/phi-4","title":"OpenRouter model listing"},{"url":"https://huggingface.co/microsoft/phi-4","title":"Microsoft Hugging Face model card"}],"display_name":"Phi-4","context_window":16000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.125,"output_per_mtok_usd":0.5}},{"id":"deepseek/deepseek-v4-flash-0731","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":37.7,"artificial_analysis_coding_index":69.1,"artificial_analysis_agentic_index":41,"artificial_analysis_economics_index":40.3,"artificial_analysis_engineering_index":34.9,"artificial_analysis_intelligence_index":34.3,"artificial_analysis_strategy_and_ops_index":43.1,"artificial_analysis_finance_and_accounting_index":37.5,"artificial_analysis_healthcare_and_medical_index":35.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.385542168674699,"lcr":0.796666666666667,"gpqa":0.908080808080808,"tau2":null,"aime25":null,"critpt":0.165714285714286,"gdpval":1433.47,"ifbench":null,"mmmuPro":null,"scicode":0.503472222222222,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":284,"tauBanking":0.393814432989691,"healthbench":null,"omniscience":-14.283333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.023,"gdpPdfAllPass":0.11,"livecodebench":null,"frontierCode20":null,"hostModelCount":18,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.786516853932584,"terminalBench40":0.121212121212121,"critptAgenticHard":null,"intelligenceIndex":34.3305129347816,"terminalbenchHard":null,"briefcaseTotalCost":25.81167335418383,"price1mInputTokens":0.14,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.4038333333333333,"price1mOutputTokens":0.28,"pricePer1k1mpImages":0.03604,"terminalBenchScience":null,"price1mBlended0To1To1":0.21000000000000002,"price1mBlended0To3To1":0.17500000000000002,"price1mBlended7To2To1":0.07210000000000001,"cacheHitDiscountPercent":0.8,"healthbenchProfessional":null,"price1mBlended0To100To1":0.1413861386138614,"price1mBlended100To1To1":0.026666666666666665,"automationBenchPartialScore":0.5396771855133674,"intelligenceIndexTimePerTask":417.67064200216106,"omniscienceHallucinationRate":0.9169695275370422,"inferenceParametersActiveBillions":13},"name":"DeepSeek V4 Flash 0731 (Max)","slug":"deepseek-v4-flash","source":"https://artificialanalysis.ai/models/deepseek-v4-flash","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v4-flash-0731/benchmarks"},"created":0,"released_at":"2026-07-31","release_source":"https://api-docs.deepseek.com/updates/","description":"DeepSeek V4 Flash 0731 is the dated snapshot of DeepSeek's efficiency-focused V4 Flash mixture-of-experts model, positioned for coding, reasoning, and agent workflows. It is the official release that followed the V4 preview, keeping the same architecture and size but re-post-trained, with native support for the Responses API format.","description_sources":[{"url":"https://openrouter.ai/deepseek/deepseek-v4-flash-0731","title":"OpenRouter model listing"},{"url":"https://api-docs.deepseek.com/updates/","title":"DeepSeek API changelog"}],"display_name":"DeepSeek V4 Flash 0731","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.14,"output_per_mtok_usd":0.28,"cached_input_per_mtok_usd":0.028}},{"id":"deepseek/deepseek-v4-pro-0813","providers":["deepseek"],"provider_details":[{"provider":"deepseek","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":41.1,"artificial_analysis_coding_index":68.8,"artificial_analysis_agentic_index":41.3,"artificial_analysis_economics_index":43.7,"artificial_analysis_engineering_index":37.1,"artificial_analysis_intelligence_index":36,"artificial_analysis_strategy_and_ops_index":46.5,"artificial_analysis_finance_and_accounting_index":41.4,"artificial_analysis_healthcare_and_medical_index":38.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"modelCount":2,"intelligenceIndex":35.9967791278402},"name":"DeepSeek V4 Pro 0813","slug":"deepseek-v4-pro","source":"https://artificialanalysis.ai/models/deepseek-v4-pro","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v4-pro-0813/benchmarks"},"created":0,"released_at":"2026-08-13","release_source":"https://api-docs.deepseek.com/updates/","description":"DeepSeek V4 Pro 0813 is the dated general-availability snapshot of DeepSeek V4 Pro, the large mixture-of-experts flagship for reasoning, coding, and agents. Compared with the preview it brings significantly enhanced agent capabilities, selectable thinking-effort levels, and native Responses API support.","description_sources":[{"url":"https://openrouter.ai/deepseek/deepseek-v4-pro-0813","title":"OpenRouter model listing"},{"url":"https://api-docs.deepseek.com/updates/","title":"DeepSeek API changelog"}],"display_name":"DeepSeek V4 Pro 0813","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.74,"output_per_mtok_usd":3.48,"cached_input_per_mtok_usd":0.145}},{"id":"deepseek/deepseek-v4.1-flash","providers":["azure-foundry","deepseek","fireworks"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"deepseek","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":41.3,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":44.4,"artificial_analysis_engineering_index":39.3,"artificial_analysis_intelligence_index":39.5,"artificial_analysis_strategy_and_ops_index":51.7,"artificial_analysis_finance_and_accounting_index":45,"artificial_analysis_healthcare_and_medical_index":40.6}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.392493049119555,"lcr":0.84,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.142857142857143,"gdpval":1600,"ifbench":null,"mmmuPro":0.769942196531792,"scicode":0.518518518518518,"harveyLab":0.016666666666666666,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.469161958568738,"parameters":552,"tauBanking":null,"healthbench":0.6088631027372047,"mlcrOverall":0.238888888888889,"omniscience":-5.3,"analystAgent":0.35,"frontierCode":0.440335333333333,"cacheHitPrice":0.006,"code2videoElo":1053.11,"gdpPdfAllPass":0.128,"livecodebench":null,"frontierCode20":null,"hostModelCount":20,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.267676767676768,"critptAgenticHard":null,"intelligenceIndex":39.456167472527,"terminalbenchHard":null,"briefcaseTotalCost":61.854780631550526,"price1mInputTokens":0.3,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.464,"price1mOutputTokens":1.2,"pricePer1k1mpImages":0.1836,"terminalBenchScience":0.0904761904761905,"price1mBlended0To1To1":0.75,"price1mBlended0To3To1":0.5249999999999999,"price1mBlended7To2To1":0.1842,"cacheHitDiscountPercent":0.9766666666666667,"healthbenchProfessional":0.506964430189633,"price1mBlended0To100To1":0.3089108910891089,"price1mBlended100To1To1":0.020588235294117643,"automationBenchPartialScore":0.6889097769674057,"intelligenceIndexTimePerTask":372.7026065605167,"omniscienceHallucinationRate":0.9645522388059702,"inferenceParametersActiveBillions":16},"name":"DeepSeek V4.1 Flash (Max)","slug":"deepseek-v4-1-flash","source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v4.1-flash/benchmarks"},"created":0,"released_at":"2026-09-10","release_source":"https://api-docs.deepseek.com/news/news260910/","description":"DeepSeek V4.1 Flash is the smallest model in DeepSeek's new architecture family, a sparse mixture-of-experts model built for coding, reasoning, and agent workloads. It is the first DeepSeek model on the Causal Encoder-Decoder design, adds native visual understanding, and sharply cuts KV-cache requirements for faster, cheaper agent runs.","description_sources":[{"url":"https://openrouter.ai/deepseek/deepseek-v4.1-flash","title":"OpenRouter model listing"},{"url":"https://api-docs.deepseek.com/news/news260910/","title":"DeepSeek release announcement"}],"display_name":"DeepSeek V4.1 Flash","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.3,"output_per_mtok_usd":1.2,"cached_input_per_mtok_usd":0.006}},{"id":"deepseek/deepseek-v4.1-flash-1017","providers":["deepseek","fireworks"],"provider_details":[{"provider":"deepseek","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":1048576,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":46.4,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":50.5,"artificial_analysis_engineering_index":45.2,"artificial_analysis_intelligence_index":45.1,"artificial_analysis_strategy_and_ops_index":52.5,"artificial_analysis_finance_and_accounting_index":48.3,"artificial_analysis_healthcare_and_medical_index":43.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-27T01:31:55.245Z"},"data":{"hle":0.433271547729379,"lcr":0.826666666666667,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.24,"gdpval":1745.77,"ifbench":null,"mmmuPro":0.80635838150289,"scicode":0.559027777777778,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":552,"tauBanking":null,"mlcrOverall":0.266666666666667,"omniscience":4.7,"analystAgent":null,"frontierCode":0.436605666666667,"cacheHitPrice":0.006,"gdpPdfAllPass":0.172,"livecodebench":null,"hostModelCount":22,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.338383838383838,"critptAgenticHard":null,"intelligenceIndex":45.0965048681912,"terminalbenchHard":null,"briefcaseTotalCost":61.72396166296012,"price1mInputTokens":0.3,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.4285,"price1mOutputTokens":1.2,"pricePer1k1mpImages":null,"terminalBenchScience":0.128571428571429,"price1mBlended0To1To1":0.75,"price1mBlended0To3To1":0.5249999999999999,"price1mBlended7To2To1":0.1842,"cacheHitDiscountPercent":0.98,"price1mBlended0To100To1":0.3089108910891089,"price1mBlended100To1To1":0.020588235294117643,"automationBenchPartialScore":0.6664095364949257,"intelligenceIndexTimePerTask":392.1812903348369,"omniscienceHallucinationRate":0.6675415573053368,"inferenceParametersActiveBillions":16},"name":"DeepSeek V4.1 Flash 1017 (Max)","slug":"deepseek-v4-1-flash-1017","source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash-1017","fetchedAt":"2026-10-27T01:31:55.245Z","pageRecords":[]}],"checkedAt":"2026-10-27T01:31:55.245Z","fetchedAt":"2026-10-27T01:31:55.245Z","nextCheckAt":"2026-11-03T01:31:55.245Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v4.1-flash-1017/benchmarks"},"created":0,"released_at":"2026-10-17","release_source":"https://api-docs.deepseek.com/news/news261017/","description":"DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, a sparse mixture-of-experts model with native image understanding and a compressed KV cache. It is the cost-efficient tier of the V4.1 family, tuned over the previous Flash snapshot for coding, terminal and long-horizon agent work.","description_sources":[{"url":"https://api-docs.deepseek.com/news/news261017/","title":"DeepSeek release announcement"},{"url":"https://openrouter.ai/deepseek/deepseek-v4.1-flash-1017","title":"OpenRouter: DeepSeek V4.1 Flash 1017"},{"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017","title":"Hugging Face: DeepSeek-V4.1-Flash-1017"}],"display_name":"DeepSeek V4.1 Flash 1017","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.3,"output_per_mtok_usd":1.2,"cached_input_per_mtok_usd":0.006}},{"id":"nvidia/nemotron-3.5-lightning","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":15.5,"artificial_analysis_coding_index":26.8,"artificial_analysis_agentic_index":3.5,"artificial_analysis_economics_index":22.8,"artificial_analysis_engineering_index":18.1,"artificial_analysis_intelligence_index":12.9,"artificial_analysis_strategy_and_ops_index":6.9,"artificial_analysis_finance_and_accounting_index":12.2,"artificial_analysis_healthcare_and_medical_index":11.5}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.105653382761817,"lcr":0.603333333333333,"gpqa":0.743434343434343,"tau2":null,"aime25":null,"critpt":0,"gdpval":637.72,"suffix":null,"ifbench":null,"mmmuPro":null,"scicode":0.320601851851852,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":31.6,"tauBanking":0.088659793814433,"healthbench":null,"mlcrOverall":0.0222222222222222,"omniscience":-17.716666666666665,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.035,"gdpPdfAllPass":0.03,"livecodebench":null,"frontierCode20":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.243445692883895,"terminalBench40":0.00505050505050505,"enterpriseOpsGym":0.18352730528200537,"critptAgenticHard":null,"intelligenceIndex":12.8572306217144,"terminalbenchHard":null,"briefcaseTotalCost":25.803993233540318,"price1mInputTokens":0.06,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.14433333333333334,"price1mOutputTokens":0.2,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.13,"price1mBlended0To3To1":0.095,"price1mBlended7To2To1":0.056499999999999995,"cacheHitDiscountPercent":0.6142857142857143,"healthbenchProfessional":null,"price1mBlended0To100To1":0.061386138613861385,"price1mBlended100To1To1":0.036862745098039225,"automationBenchPartialScore":0.008410069634719542,"intelligenceIndexTimePerTask":103.6392953599501,"omniscienceHallucinationRate":0.3757304246201792,"inferenceParametersActiveBillions":3.6},"name":"Nemotron 3.5 Lightning","slug":"nemotron-3-5-lightning","source":"https://artificialanalysis.ai/models/nemotron-3-5-lightning","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/nvidia/nemotron-3.5-lightning/benchmarks"},"created":0,"released_at":"2026-08-11","release_source":"https://blogs.nvidia.com/blog/nemotron-lightning-switchyard-rtx-dgx/","description":"Nemotron 3.5 Lightning is NVIDIA's open mixture-of-experts model, the efficiency-focused member of the Nemotron 3 family, built for high-throughput, always-on agentic workloads. It targets specialized tasks such as code review, tool use, and alert monitoring, runs locally on RTX PCs, DGX Spark, and Jetson, and can be post-trained with NeMo.","description_sources":[{"url":"https://openrouter.ai/nvidia/nemotron-3.5-lightning","title":"OpenRouter model listing"},{"url":"https://blogs.nvidia.com/blog/nemotron-lightning-switchyard-rtx-dgx/","title":"NVIDIA blog announcement"}],"display_name":"Nemotron 3.5 Lightning","context_window":262144,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.05,"output_per_mtok_usd":0.2,"cached_input_per_mtok_usd":0.01}},{"id":"deepseek/deepseek-v4-flash","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":37.7,"artificial_analysis_coding_index":69.1,"artificial_analysis_agentic_index":41,"artificial_analysis_economics_index":40.3,"artificial_analysis_engineering_index":34.9,"artificial_analysis_intelligence_index":34.3,"artificial_analysis_strategy_and_ops_index":43.1,"artificial_analysis_finance_and_accounting_index":37.5,"artificial_analysis_healthcare_and_medical_index":35.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.385542168674699,"lcr":0.796666666666667,"gpqa":0.908080808080808,"tau2":null,"aime25":null,"critpt":0.165714285714286,"gdpval":1433.47,"ifbench":null,"mmmuPro":null,"scicode":0.503472222222222,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":284,"tauBanking":0.393814432989691,"healthbench":null,"omniscience":-14.283333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.023,"gdpPdfAllPass":0.11,"livecodebench":null,"frontierCode20":null,"hostModelCount":18,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.786516853932584,"terminalBench40":0.121212121212121,"critptAgenticHard":null,"intelligenceIndex":34.3305129347816,"terminalbenchHard":null,"briefcaseTotalCost":25.81167335418383,"price1mInputTokens":0.14,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.4038333333333333,"price1mOutputTokens":0.28,"pricePer1k1mpImages":0.03604,"terminalBenchScience":null,"price1mBlended0To1To1":0.21000000000000002,"price1mBlended0To3To1":0.17500000000000002,"price1mBlended7To2To1":0.07210000000000001,"cacheHitDiscountPercent":0.8,"healthbenchProfessional":null,"price1mBlended0To100To1":0.1413861386138614,"price1mBlended100To1To1":0.026666666666666665,"automationBenchPartialScore":0.5396771855133674,"intelligenceIndexTimePerTask":417.67064200216106,"omniscienceHallucinationRate":0.9169695275370422,"inferenceParametersActiveBillions":13},"name":"DeepSeek V4 Flash 0731 (Max)","slug":"deepseek-v4-flash","source":"https://artificialanalysis.ai/models/deepseek-v4-flash","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v4-flash/benchmarks"},"created":0,"released_at":"2026-04-24","release_source":"https://api-docs.deepseek.com/updates/","description":"DeepSeek V4 Flash is DeepSeek's efficiency-optimized mixture-of-experts model, designed for fast inference on coding, reasoning, and agent workloads. It pairs a large total parameter count with a small active set, supports long inputs, and later gained an experimental vision variant before being superseded by V4.1 Flash.","description_sources":[{"url":"https://openrouter.ai/deepseek/deepseek-v4-flash","title":"OpenRouter model listing"},{"url":"https://api-docs.deepseek.com/updates/","title":"DeepSeek API changelog"}],"display_name":"DeepSeek V4 Flash","context_window":1000000,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.19,"output_per_mtok_usd":0.51,"cached_input_per_mtok_usd":0.028}},{"id":"deepseek/deepseek-v4-pro","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":41.1,"artificial_analysis_coding_index":68.8,"artificial_analysis_agentic_index":41.3,"artificial_analysis_economics_index":43.7,"artificial_analysis_engineering_index":37.1,"artificial_analysis_intelligence_index":36,"artificial_analysis_strategy_and_ops_index":46.5,"artificial_analysis_finance_and_accounting_index":41.4,"artificial_analysis_healthcare_and_medical_index":38.2}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"modelCount":2,"intelligenceIndex":35.9967791278402},"name":"DeepSeek V4 Pro 0813","slug":"deepseek-v4-pro","source":"https://artificialanalysis.ai/models/deepseek-v4-pro","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v4-pro/benchmarks"},"created":0,"released_at":"2026-04-24","release_source":"https://api-docs.deepseek.com/updates/","description":"DeepSeek V4 Pro is the large mixture-of-experts flagship of DeepSeek's V4 generation, built for advanced reasoning, coding, and production agent systems. It offers low, high, and max thinking-effort levels, native tool use through the Responses API format, and long-context inputs.","description_sources":[{"url":"https://openrouter.ai/deepseek/deepseek-v4-pro","title":"OpenRouter model listing"},{"url":"https://api-docs.deepseek.com/updates/","title":"DeepSeek API changelog"}],"display_name":"DeepSeek V4 Pro","context_window":1000000,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.74,"output_per_mtok_usd":3.48,"cached_input_per_mtok_usd":0.145}},{"id":"deepseek/deepseek-v3-2","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":16,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.112140871177016,"lcr":0.456666666666667,"gpqa":0.750505050505051,"tau2":0.789473684210526,"aime25":0.59,"critpt":0.00857142857142857,"gdpval":null,"ifbench":0.490476190476191,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":685,"tauBanking":null,"healthbench":null,"omniscience":-46.88333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":0.592592592592593,"frontierCode20":null,"hostModelCount":10,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"critptAgenticHard":null,"intelligenceIndex":16.043537719683,"terminalbenchHard":0.325757575757576,"price1mInputTokens":0.28,"contextWindowTokens":128000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.24,"price1mOutputTokens":0.42,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.35,"price1mBlended0To3To1":0.315,"price1mBlended7To2To1":0.29400000000000004,"cacheHitDiscountPercent":null,"healthbenchProfessional":null,"price1mBlended0To100To1":0.2813861386138614,"price1mBlended100To1To1":0.2813725490196079,"omniscienceHallucinationRate":0.9326754385964913,"inferenceParametersActiveBillions":37},"name":"DeepSeek V3.2 (Non-reasoning)","slug":"deepseek-v3-2","source":"https://artificialanalysis.ai/models/deepseek-v3-2","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v3-2/benchmarks"},"created":0,"released_at":"2025-12-01","release_source":"https://api-docs.deepseek.com/updates/","description":"DeepSeek V3.2 is DeepSeek's reasoning-first general model built for agents, serving both the chat and reasoner endpoints in non-thinking and thinking modes. It introduces DeepSeek Sparse Attention for efficient long-context work and integrates thinking directly into tool use in both modes.","description_sources":[{"url":"https://openrouter.ai/deepseek/deepseek-v3.2","title":"OpenRouter model listing"},{"url":"https://api-docs.deepseek.com/updates/","title":"DeepSeek API changelog"}],"display_name":"DeepSeek V3.2","context_window":163840,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.58,"output_per_mtok_usd":1.68}},{"id":"deepseek/deepseek-v3-2-speciale","providers":["azure-foundry"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":14.5,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.287303058387396,"lcr":0.7,"gpqa":0.870707070707071,"tau2":0,"aime25":0.966666666666667,"critpt":0.0742857142857143,"gdpval":null,"suffix":null,"ifbench":0.638775510204082,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":685,"tauBanking":null,"healthbench":null,"omniscience":-17.716666666666665,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":0.896296296296296,"frontierCode20":null,"hostModelCount":0,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":null,"critptAgenticHard":null,"intelligenceIndex":14.4560679389612,"terminalbenchHard":0.348484848484849,"price1mInputTokens":null,"contextWindowTokens":128000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.37766666666666665,"price1mOutputTokens":null,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":null,"price1mBlended0To3To1":null,"price1mBlended7To2To1":null,"cacheHitDiscountPercent":null,"healthbenchProfessional":null,"price1mBlended0To100To1":null,"price1mBlended100To1To1":null,"omniscienceHallucinationRate":0.8915372254954472,"inferenceParametersActiveBillions":37},"name":"DeepSeek V3.2 Speciale","slug":"deepseek-v3-2-speciale","source":"https://artificialanalysis.ai/models/deepseek-v3-2-speciale","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/deepseek/deepseek-v3-2-speciale/benchmarks"},"created":0,"released_at":"2025-12-01","release_source":"https://api-docs.deepseek.com/updates/","description":"DeepSeek V3.2 Speciale is a temporary, API-only variant of DeepSeek V3.2 with maxed-out reasoning, aimed at hard mathematical and competitive-programming problems. Unlike the standard V3.2 it does not support tool calls and was served through a dedicated endpoint for a limited window.","description_sources":[{"url":"https://api-docs.deepseek.com/news/news251201","title":"DeepSeek release announcement"},{"url":"https://api-docs.deepseek.com/updates/","title":"DeepSeek API changelog"}],"display_name":"DeepSeek V3.2 Speciale","context_window":163840,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.58,"output_per_mtok_usd":1.68}},{"id":"zai/glm-5.3","providers":["azure-foundry","zai","fireworks"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"northcentralus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":42,"artificial_analysis_coding_index":74.8,"artificial_analysis_agentic_index":53.1,"artificial_analysis_economics_index":51.2,"artificial_analysis_engineering_index":47.1,"artificial_analysis_intelligence_index":44.8,"artificial_analysis_strategy_and_ops_index":47.6,"artificial_analysis_finance_and_accounting_index":44.7,"artificial_analysis_healthcare_and_medical_index":46.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"modelCount":2,"intelligenceIndex":44.777392385614},"name":"GLM-5.3","slug":"glm-5-3","source":"https://artificialanalysis.ai/models/glm-5-3","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/zai/glm-5.3/benchmarks"},"created":0,"released_at":"2026-08-14","release_source":"https://z.ai/blog/glm-5.3","description":"GLM-5.3 is Z.AI's flagship reasoning model, built for complex software engineering, long-horizon agent tasks, and security work such as vulnerability discovery. It succeeds GLM-5.2 with a large gain on Z.AI's own coding bench and supports long text inputs with tool calling.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-5.3","title":"OpenRouter model listing"},{"url":"https://z.ai/blog/glm-5.3","title":"Z.AI release announcement"}],"display_name":"GLM-5.3","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.4,"output_per_mtok_usd":4.4,"cached_input_per_mtok_usd":0.26}},{"id":"zai/glm-5.3-fast","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:07:54.046Z","nextCheckAt":"2026-10-28T20:07:54.046Z","summary":true,"detailPath":"/v1/models/zai/glm-5.3-fast/benchmarks"},"created":0,"released_at":"2026-08-14","release_source":"https://z.ai/blog/glm-5.3","description":"GLM-5.3 Fast is a higher-throughput serving variant of Z.AI's GLM-5.3, using the same weights so output quality matches the standard endpoint. It targets agent loops that chain many model calls, keeping GLM-5.3's always-on thinking, reasoning-effort control, tool calling, and long-context inputs at faster speeds.","description_sources":[{"url":"https://docs.fireworks.ai/serverless/pricing","title":"Fireworks serverless pricing"},{"url":"https://z.ai/blog/glm-5.3","title":"Z.AI release announcement"}],"display_name":"GLM-5.3 Fast","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":2.1,"output_per_mtok_usd":6.6,"cached_input_per_mtok_usd":0.39}},{"id":"zai/glm-5.3-flash","providers":["azure-foundry","zai","fireworks"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":39.9,"artificial_analysis_coding_index":71.5,"artificial_analysis_agentic_index":50.9,"artificial_analysis_economics_index":48.9,"artificial_analysis_engineering_index":43.8,"artificial_analysis_intelligence_index":41.8,"artificial_analysis_strategy_and_ops_index":45.1,"artificial_analysis_finance_and_accounting_index":42,"artificial_analysis_healthcare_and_medical_index":44.8}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.398517145505097,"lcr":0.8,"gpqa":0.912121212121212,"tau2":null,"aime25":null,"critpt":0.154285714285714,"gdpval":1650.37,"suffix":null,"aa3dElo":936.6148,"ifbench":null,"mmmuPro":null,"scicode":0.516203703703704,"harveyLab":0.008333333333333333,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.512429378531074,"parameters":320,"tauBanking":0.472164948453608,"healthbench":null,"mlcrOverall":0.116666666666667,"omniscience":7.466666666666667,"analystAgent":0.25,"frontierCode":0.409957333333333,"cacheHitPrice":0.03,"code2videoElo":974.54,"gdpPdfAllPass":0.154,"livecodebench":null,"frontierCode20":null,"hostModelCount":26,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.842696629213483,"terminalBench40":0.328282828282828,"enterpriseOpsGym":0.3321396598030439,"critptAgenticHard":null,"intelligenceIndex":41.807466113455,"terminalbenchHard":null,"briefcaseTotalCost":89.54268436893602,"price1mInputTokens":0.15,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.275,"price1mOutputTokens":0.5,"pricePer1k1mpImages":null,"terminalBenchScience":0.0476190476190476,"price1mBlended0To1To1":0.325,"price1mBlended0To3To1":0.2375,"price1mBlended7To2To1":0.101,"cacheHitDiscountPercent":0.8,"healthbenchProfessional":null,"price1mBlended0To100To1":0.15346534653465346,"price1mBlended100To1To1":0.035784313725490194,"automationBenchPartialScore":0.6036862782167782,"intelligenceIndexTimePerTask":721.9145911593189,"omniscienceHallucinationRate":0.27632183908045976,"inferenceParametersActiveBillions":18},"name":"GLM-5.3-Flash","slug":"glm-5-3-flash","source":"https://artificialanalysis.ai/models/glm-5-3-flash","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/zai/glm-5.3-flash/benchmarks"},"created":0,"released_at":"2026-08-26","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-5.3-Flash is the smaller, efficient mixture-of-experts sibling of Z.AI's GLM-5.3, positioned for coding, office documents, and financial research with autonomous tool use. It adds native visual understanding of interfaces and rendered output, using a hybrid sparse and linear attention design for accurate long-context behavior.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-5.3-flash","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-5.3-Flash","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.15,"output_per_mtok_usd":0.5,"cached_input_per_mtok_usd":0.03}},{"id":"zai/glm-5.3-flashx","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:07:54.046Z","nextCheckAt":"2026-10-28T20:07:54.046Z","summary":true,"detailPath":"/v1/models/zai/glm-5.3-flashx/benchmarks"},"created":0,"description":"GLM-5.3-FlashX is the high-speed serving option for Z.AI's natively multimodal GLM-5.3-Flash weights, aimed at latency-sensitive coding and agent loops. It keeps the same always-on thinking, tool calling, and visual understanding of interfaces and rendered output as its Flash sibling, trading a higher rate for faster responses.","description_sources":[{"url":"https://docs.z.ai/guides/vlm/glm-5.3-flash","title":"Z.AI GLM-5.3-Flash/FlashX model documentation"},{"url":"https://vercel.com/ai-gateway/models/glm-5.3-flashx","title":"Vercel AI Gateway model listing"}],"display_name":"GLM-5.3-FlashX","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.37,"output_per_mtok_usd":1.25,"cached_input_per_mtok_usd":0.075}},{"id":"zai/glm-5.2","providers":["azure-foundry","zai"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":34.3,"artificial_analysis_coding_index":68.8,"artificial_analysis_agentic_index":38.4,"artificial_analysis_economics_index":45.6,"artificial_analysis_engineering_index":37.2,"artificial_analysis_intelligence_index":33.7,"artificial_analysis_strategy_and_ops_index":30.6,"artificial_analysis_finance_and_accounting_index":33.3,"artificial_analysis_healthcare_and_medical_index":32.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T16:24:48.193Z"},"data":{"hle":0.411492122335496,"lcr":0.783333333333333,"gpqa":0.894949494949495,"tau2":0.991228070175439,"aime25":null,"critpt":0.208571428571429,"gdpval":1373.04,"ifbench":0.733333333333333,"mmmuPro":null,"scicode":0.511574074074074,"apexAgents":0.337020648967552,"hleDiamond":null,"itBenchSre":0.426553672316384,"parameters":753,"tauBanking":0.34639175257732,"healthbench":null,"mlcrOverall":0.127777777777778,"omniscience":4.433333333333334,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.26,"gdpPdfAllPass":0.104,"livecodebench":null,"frontierCode20":null,"hostModelCount":19,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.779026217228464,"terminalBench40":0.0101010101010101,"enterpriseOpsGym":0.42733512384362876,"critptAgenticHard":null,"intelligenceIndex":33.7054572377545,"terminalbenchHard":0.507575757575758,"briefcaseTotalCost":224.19185994965505,"price1mInputTokens":1.4,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.24333333333333335,"price1mOutputTokens":4.4,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":2.9000000000000004,"price1mBlended0To3To1":2.15,"price1mBlended7To2To1":0.9019999999999999,"cacheHitDiscountPercent":0.8142857142857143,"healthbenchProfessional":null,"price1mBlended0To100To1":1.4297029702970299,"price1mBlended100To1To1":0.3117647058823529,"automationBenchPartialScore":0.2839897721005435,"intelligenceIndexTimePerTask":574.2402130035053,"omniscienceHallucinationRate":0.2629955947136564,"inferenceParametersActiveBillions":40},"name":"GLM-5.2 (Max)","slug":"glm-5-2","source":"https://artificialanalysis.ai/models/glm-5-2","fetchedAt":"2026-10-27T18:34:54.689Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:34:54.689Z","fetchedAt":"2026-10-14T16:24:48.193Z","nextCheckAt":"2026-10-28T18:34:54.689Z","summary":true,"detailPath":"/v1/models/zai/glm-5.2/benchmarks"},"created":0,"released_at":"2026-06-16","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-5.2 is Z.AI's flagship foundation model for long-horizon engineering, from project-level codebase takeover and large refactors to deep debugging and research reproduction. It supports lossless long-context inputs, multiple thinking modes, streaming, and function calling, and Z.AI positions it as its leading open-source coding model.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-5.2","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-5.2","context_window":1000000,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.4,"output_per_mtok_usd":4.4,"cached_input_per_mtok_usd":0.26}},{"id":"zai/glm-5.1","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":28.8,"artificial_analysis_coding_index":55.8,"artificial_analysis_agentic_index":23.9,"artificial_analysis_economics_index":37.6,"artificial_analysis_engineering_index":30.2,"artificial_analysis_intelligence_index":26.1,"artificial_analysis_strategy_and_ops_index":23.3,"artificial_analysis_finance_and_accounting_index":26.3,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.300741427247451,"lcr":0.736666666666667,"gpqa":0.867676767676768,"tau2":0.976608187134503,"aime25":null,"critpt":0.0457142857142857,"gdpval":1116.24,"ifbench":0.762585034013605,"mmmuPro":null,"scicode":0.447916666666667,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.402542372881356,"parameters":744,"tauBanking":0.136082474226804,"omniscience":0.85,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.26,"gdpPdfAllPass":0.084,"livecodebench":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":3791,"terminalBench21":0.617977528089888,"terminalBench40":0.0202020202020202,"gdpvalNormalized":0.30812,"intelligenceIndex":26.0585912980095,"terminalbenchHard":0.431818181818182,"briefcaseTotalCost":76.41437041545339,"price1mInputTokens":1.285,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.237,"price1mOutputTokens":4.07,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":2.6775,"price1mBlended0To3To1":1.98125,"price1mBlended7To2To1":0.8460000000000001,"cacheHitDiscountPercent":0.8115942028985507,"price1mBlended0To100To1":1.3125742574257424,"price1mBlended100To1To1":0.30740196078431375,"automationBenchPartialScore":0.20308354687857075,"intelligenceIndexTimePerTask":666.5466318374492,"omniscienceHallucinationRate":0.299475753604194,"inferenceParametersActiveBillions":40},"name":"GLM-5.1 (Reasoning)","slug":"glm-5-1","source":"https://artificialanalysis.ai/models/glm-5-1","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/zai/glm-5.1/benchmarks"},"created":0,"released_at":"2026-04-07","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-5.1 is Z.AI's flagship coding model, built for long-horizon software tasks that require autonomous planning, bug fixing, and iterative refinement. Where earlier GLM models were designed around minute-level interactions, GLM-5.1 can work independently and continuously for hours in a single run.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-5.1","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-5.1","context_window":200000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.4,"output_per_mtok_usd":4.4,"cached_input_per_mtok_usd":0.26}},{"id":"zai/glm-5","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":27.9,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.292863762743281,"lcr":0.756666666666667,"gpqa":0.82020202020202,"tau2":0.982456140350877,"aime25":null,"critpt":0.02,"gdpval":null,"ifbench":0.72312925170068,"mmmuPro":null,"scicode":null,"apexAgents":0.14454277286135694,"hleDiamond":null,"itBenchSre":null,"parameters":744,"tauBanking":null,"omniscience":0.26666666666666666,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"livecodebench":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":3105,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":27.9111550817227,"terminalbenchHard":0.431818181818182,"price1mInputTokens":1,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.263,"price1mOutputTokens":3.2,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":2.1,"price1mBlended0To3To1":1.55,"price1mBlended7To2To1":0.66,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":1.0217821782178218,"price1mBlended100To1To1":0.2372549019607843,"omniscienceHallucinationRate":0.35323383084577115,"inferenceParametersActiveBillions":40},"name":"GLM-5 (Reasoning)","slug":"glm-5","source":"https://artificialanalysis.ai/models/glm-5","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/zai/glm-5/benchmarks"},"created":0,"released_at":"2026-02-11","release_source":"https://modal.com/blog/try-glm-5","description":"GLM-5 is Z.AI's flagship open-source foundation model, engineered for complex systems design and long-horizon agent workflows aimed at expert developers. It shifts the family's focus from coding to engineering, targeting backend architecture, complex algorithms, and stubborn debugging on large-scale programming tasks.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-5","title":"OpenRouter model listing"},{"url":"https://modal.com/blog/try-glm-5","title":"Modal launch write-up for GLM-5"}],"display_name":"GLM-5","context_window":200000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":3.2,"cached_input_per_mtok_usd":0.2}},{"id":"zai/glm-5-turbo","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":26.6,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.277571825764597,"lcr":0.716666666666667,"gpqa":0.847474747474747,"tau2":0.985380116959064,"aime25":null,"critpt":0.00285714285714286,"gdpval":null,"suffix":null,"ifbench":0.731972789115646,"mmmuPro":null,"scicode":null,"sizeClass":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":-16.366666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":null,"hostModelCount":0,"cacheWritePrice":null,"reasoningTokens":2417,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":26.5965803042835,"terminalbenchHard":0.333333333333333,"price1mInputTokens":null,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.2845,"price1mOutputTokens":null,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":null,"price1mBlended0To3To1":null,"price1mBlended7To2To1":null,"cacheHitDiscountPercent":null,"price1mBlended0To100To1":null,"price1mBlended100To1To1":null,"omniscienceHallucinationRate":0.6263685068716516,"inferenceParametersActiveBillions":null},"name":"GLM-5-Turbo","slug":"glm-5-turbo","source":"https://artificialanalysis.ai/models/glm-5-turbo","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/zai/glm-5-turbo/benchmarks"},"created":0,"released_at":"2026-03-15","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-5 Turbo is a fast-inference member of Z.AI's GLM-5 family, tuned for agent-driven environments such as OpenClaw. It is deeply optimized for real-world agent workflows where response speed matters, and Z.AI bundles it in its GLM Coding Plan alongside GLM-5.3 and GLM-5.2.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-5-turbo","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-5 Turbo","context_window":200000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.2,"output_per_mtok_usd":4,"cached_input_per_mtok_usd":0.24}},{"id":"zai/glm-4.7","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":45.3,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":22.2,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.27386468952734,"lcr":0.71,"gpqa":0.858585858585858,"tau2":0.95906432748538,"aime25":0.95,"critpt":0.0171020408163265,"gdpval":1014.22,"ifbench":0.67891156462585,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":357,"tauBanking":0.121649484536082,"omniscience":-36.43333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.45,"livecodebench":0.894179894179894,"hostModelCount":5,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.453183520599251,"terminalBench40":null,"gdpvalNormalized":0.25711,"intelligenceIndex":22.2402659332875,"terminalbenchHard":0.318181818181818,"price1mInputTokens":0.6,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.2931666666666667,"price1mOutputTokens":2.2,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":1.4000000000000001,"price1mBlended0To3To1":1,"price1mBlended7To2To1":0.655,"cacheHitDiscountPercent":0.8083333333333333,"price1mBlended0To100To1":0.6158415841584158,"price1mBlended100To1To1":0.4686274509803922,"omniscienceHallucinationRate":0.9302051402970998,"inferenceParametersActiveBillions":32},"name":"GLM-4.7 (Reasoning)","slug":"glm-4-7","source":"https://artificialanalysis.ai/models/glm-4-7","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/zai/glm-4.7/benchmarks"},"created":0,"released_at":"2025-12-22","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-4.7 is Z.AI's flagship foundation model of the GLM-4 generation, upgraded for programming, reasoning, and agentic work. Compared with GLM-4.6 it brings stronger coding and more stable multi-step reasoning and execution, with improvements in real-world development workflows, front-end generation, and document tasks.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-4.7","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-4.7","context_window":200000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.6,"output_per_mtok_usd":2.2,"cached_input_per_mtok_usd":0.11}},{"id":"zai/glm-4.6","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":14.9,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.0546802594995366,"lcr":0.263333333333333,"gpqa":0.632323232323232,"tau2":0.769005847953216,"aime25":0.443333333333333,"critpt":0,"gdpval":null,"ifbench":0.36734693877551,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":357,"tauBanking":null,"omniscience":-31.683333333333334,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.355,"livecodebench":0.560846560846561,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":0,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":14.9292992892622,"terminalbenchHard":0.287878787878788,"price1mInputTokens":0.575,"contextWindowTokens":200000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.21416666666666667,"price1mOutputTokens":2.2,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":1.3875000000000002,"price1mBlended0To3To1":0.98125,"price1mBlended7To2To1":0.5835,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.5910891089108911,"price1mBlended100To1To1":0.37524509803921574,"omniscienceHallucinationRate":0.6757158006362672,"inferenceParametersActiveBillions":32},"name":"GLM-4.6 (Non-reasoning)","slug":"glm-4-6","source":"https://artificialanalysis.ai/models/glm-4-6","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/zai/glm-4.6/benchmarks"},"created":0,"released_at":"2025-09-30","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-4.6 is Z.AI's flagship coding model in the GLM-4 line, built for programming tasks and complex agent work. Over GLM-4.5 it expands the context window, sharpens coding performance, and improves reasoning and tool use inside agent frameworks.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-4.6","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-4.6","context_window":200000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.6,"output_per_mtok_usd":2.2,"cached_input_per_mtok_usd":0.11}},{"id":"zai/glm-4.5","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["glm-4-5: AA page HTTP 404"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":12.8,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-27T20:38:55.901Z"},"data":{"hle":0.129749768303985,"lcr":0.526666666666667,"gpqa":0.781818181818182,"tau2":0.429824561403509,"aime25":0.736666666666667,"critpt":0,"gdpval":null,"ifbench":0.440816326530612,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":355,"tauBanking":null,"healthbench":null,"omniscience":-27.383333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":0.737566137566138,"frontierCode20":null,"hostModelCount":0,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":null,"critptAgenticHard":null,"intelligenceIndex":12.7680269610265,"terminalbenchHard":0.21969696969697,"price1mInputTokens":null,"contextWindowTokens":128000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.25133333333333335,"price1mOutputTokens":null,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":null,"price1mBlended0To3To1":null,"price1mBlended7To2To1":null,"cacheHitDiscountPercent":null,"healthbenchProfessional":null,"price1mBlended0To100To1":null,"price1mBlended100To1To1":null,"omniscienceHallucinationRate":0.7014692787177204,"inferenceParametersActiveBillions":32},"name":"GLM-4.5 (Reasoning)","slug":"glm-4.5","source":"https://artificialanalysis.ai/models/glm-4.5","fetchedAt":"2026-10-27T20:38:55.901Z","pageRecords":[]}],"checkedAt":"2026-10-27T20:38:55.901Z","fetchedAt":"2026-09-11T11:52:20.926Z","nextCheckAt":"2026-10-28T20:38:55.901Z","summary":true,"detailPath":"/v1/models/zai/glm-4.5/benchmarks"},"created":0,"released_at":"2025-07-28","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-4.5 is Z.AI's mixture-of-experts flagship foundation model, purpose-built for agent-based applications spanning reasoning, coding, and agentic tasks. Z.AI reports doubled parameter efficiency over prior models and one-click compatibility with the Claude Code framework.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-4.5","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-4.5","context_window":128000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.6,"output_per_mtok_usd":2.2,"cached_input_per_mtok_usd":0.11}},{"id":"zai/glm-4.5-air","providers":["zai"],"provider_details":[{"provider":"zai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":11.1,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.0704355885078777,"lcr":0.466666666666667,"gpqa":0.733333333333333,"tau2":0.464912280701754,"aime25":0.806666666666667,"critpt":0,"gdpval":null,"suffix":null,"ifbench":0.375510204081633,"mmmuPro":null,"scicode":null,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":106,"tauBanking":null,"omniscience":-61.53333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"livecodebench":0.683597883597884,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":null,"intelligenceIndex":11.0889137659922,"terminalbenchHard":0.204545454545455,"price1mInputTokens":0.17,"contextWindowTokens":128000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.1625,"price1mOutputTokens":0.98,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.575,"price1mBlended0To3To1":0.3725,"price1mBlended7To2To1":0.251,"cacheHitDiscountPercent":null,"price1mBlended0To100To1":0.178019801980198,"price1mBlended100To1To1":0.17794117647058827,"omniscienceHallucinationRate":0.9287562189054727,"inferenceParametersActiveBillions":12},"name":"GLM-4.5-Air","slug":"glm-4-5-air","source":"https://artificialanalysis.ai/models/glm-4-5-air","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/zai/glm-4.5-air/benchmarks"},"created":0,"released_at":"2025-07-28","release_source":"https://docs.z.ai/release-notes/new-released","description":"GLM-4.5 Air is the lightweight variant of Z.AI's GLM-4.5 family, also purpose-built for agent-centric applications. It keeps the mixture-of-experts architecture with a more compact parameter budget, offering a cheaper, faster option for reasoning, coding, and tool-use tasks.","description_sources":[{"url":"https://openrouter.ai/z-ai/glm-4.5-air","title":"OpenRouter model listing"},{"url":"https://docs.z.ai/release-notes/new-released","title":"Z.AI release notes"}],"display_name":"GLM-4.5 Air","context_window":128000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.2,"output_per_mtok_usd":1.1,"cached_input_per_mtok_usd":0.03}},{"id":"moonshotai/kimi-k3.1-pro-highspeed","providers":["moonshotai"],"provider_details":[{"provider":"moonshotai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T22:42:58.139Z","nextCheckAt":"2026-10-28T22:42:58.139Z","summary":true,"detailPath":"/v1/models/moonshotai/kimi-k3.1-pro-highspeed/benchmarks"},"created":0,"released_at":"2026-10-23","release_source":"https://www.kimi.com/en/blog/kimi-k3-1-pro","description":"Kimi K3.1 Pro Highspeed is the faster-serving variant of Moonshot AI's Kimi K3.1 Pro, a multimodal reasoning model built for long-horizon coding, 3D creation and agentic work. It delivers the same capabilities at roughly two to three times the output speed, suiting coding agents and interactive use where latency matters.","description_sources":[{"url":"https://www.kimi.com/en/blog/kimi-k3-1-pro","title":"Kimi K3.1 Pro announcement"},{"url":"https://platform.kimi.ai/docs/pricing/chat","title":"Kimi API model pricing"},{"url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart","title":"Kimi K3.1 Pro quickstart"},{"url":"https://vercel.com/ai-gateway/models/kimi-k3.1-pro-highspeed","title":"Vercel AI Gateway listing"},{"url":"https://www.unite.ai/moonshot-ai-releases-kimi-k3-1-pro-first-model-in-new-k3-1-series/","title":"Unite.AI release coverage"}],"display_name":"Kimi K3.1 Pro Highspeed","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":true,"default":"max"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":4,"output_per_mtok_usd":20,"cached_input_per_mtok_usd":0.2}},{"id":"moonshotai/kimi-k3.1-pro","providers":["moonshotai"],"provider_details":[{"provider":"moonshotai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":55.8,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":60,"artificial_analysis_engineering_index":55.9,"artificial_analysis_intelligence_index":55.1,"artificial_analysis_strategy_and_ops_index":59.4,"artificial_analysis_finance_and_accounting_index":56.5,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-23T19:30:16.329Z"},"data":{"hle":0.524559777571826,"lcr":0.863333333333333,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.245714285714286,"gdpval":1845.72,"suffix":null,"ifbench":null,"mmmuPro":0.838728323699422,"scicode":0.607638888888889,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":2800,"tauBanking":null,"licenseName":null,"omniscience":29.516666666666666,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"gdpPdfAllPass":0.314,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.671717171717172,"intelligenceIndex":55.0869210959395,"terminalbenchHard":null,"briefcaseTotalCost":657.0779093643412,"price1mInputTokens":2,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.5101666666666667,"price1mOutputTokens":10,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":6,"price1mBlended0To3To1":4,"price1mBlended7To2To1":1.54,"cacheHitDiscountPercent":0.9,"price1mBlended0To100To1":2.0792079207920793,"price1mBlended100To1To1":0.3137254901960784,"automationBenchPartialScore":0.7329771090921112,"intelligenceIndexTimePerTask":599.6084964990467,"omniscienceHallucinationRate":0.43892480435522285,"inferenceParametersActiveBillions":104},"name":"Kimi K3.1 Pro","slug":"kimi-k3-1-pro","source":"https://artificialanalysis.ai/models/kimi-k3-1-pro","fetchedAt":"2026-10-23T19:30:16.329Z","pageRecords":[]}],"checkedAt":"2026-10-23T19:30:16.329Z","fetchedAt":"2026-10-23T19:30:16.329Z","nextCheckAt":"2026-10-30T19:30:16.329Z","summary":true,"detailPath":"/v1/models/moonshotai/kimi-k3.1-pro/benchmarks"},"created":0,"released_at":"2026-10-23","release_source":"https://www.kimi.com/en/blog/kimi-k3-1-pro","description":"Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model, a post-trained successor to Kimi K3 built for long-horizon coding, 3D creation, financial research and agentic knowledge work. It always reasons, with a selectable effort level, and is the first model in the K3.1 series.","description_sources":[{"url":"https://www.kimi.com/en/blog/kimi-k3-1-pro","title":"Kimi K3.1 Pro announcement"},{"url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart","title":"Kimi K3.1 Pro quickstart"},{"url":"https://openrouter.ai/moonshotai/kimi-k3.1-pro","title":"OpenRouter model listing"},{"url":"https://www.unite.ai/moonshot-ai-releases-kimi-k3-1-pro-first-model-in-new-k3-1-series/","title":"Unite.AI release coverage"}],"display_name":"Kimi K3.1 Pro","context_window":1048576,"reasoning":{"kind":"effort","levels":["low","high","max"],"off":true,"default":"max"},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":10,"cached_input_per_mtok_usd":0.2}},{"id":"moonshotai/kimi-k3","providers":["azure-foundry","fireworks","moonshotai","qwen"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"moonshotai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":48.6,"artificial_analysis_coding_index":76.2,"artificial_analysis_agentic_index":50,"artificial_analysis_economics_index":53.4,"artificial_analysis_engineering_index":43.3,"artificial_analysis_intelligence_index":43.6,"artificial_analysis_strategy_and_ops_index":50.5,"artificial_analysis_finance_and_accounting_index":46.9,"artificial_analysis_healthcare_and_medical_index":45}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"modelCount":2,"intelligenceIndex":43.5938229518782},"name":"Kimi K3","slug":"kimi-k3","source":"https://artificialanalysis.ai/models/kimi-k3","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/moonshotai/kimi-k3/benchmarks"},"created":0,"released_at":"2026-07-16","release_source":"https://www.moonshot.ai/","description":"Kimi K3 is Moonshot AI's frontier open-weight multimodal reasoning model, built for long-horizon coding, knowledge work, and deep reasoning. It is natively multimodal, handles long inputs, and is positioned for agentic workflows across code, analysis, spreadsheets, and slides.","description_sources":[{"url":"https://openrouter.ai/moonshotai/kimi-k3","title":"OpenRouter model listing"},{"url":"https://www.moonshot.ai/","title":"Moonshot AI site"}],"display_name":"Kimi K3","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":3,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.3}},{"id":"qwen/qwen3.7-max","providers":["qwen"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":35.4,"artificial_analysis_coding_index":66,"artificial_analysis_agentic_index":22.5,"artificial_analysis_economics_index":43.6,"artificial_analysis_engineering_index":33.4,"artificial_analysis_intelligence_index":29.5,"artificial_analysis_strategy_and_ops_index":25.6,"artificial_analysis_finance_and_accounting_index":31.3,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.405004633920297,"lcr":0.79,"gpqa":0.923232323232323,"tau2":0.947368421052632,"aime25":null,"critpt":0.134285714285714,"gdpval":1127.65,"suffix":null,"ifbench":0.805442176870748,"mmmuPro":null,"scicode":0.49537037037037,"apexAgents":null,"hleDiamond":null,"itBenchSre":0.424670433145009,"licenseUrl":null,"parameters":null,"tauBanking":0.117525773195876,"licenseName":null,"mlcrOverall":0.144444444444444,"omniscience":13.483333333333333,"analystAgent":0.1875,"frontierCode":null,"cacheHitPrice":0.5,"gdpPdfAllPass":0.086,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":2409,"terminalBench21":0.745318352059925,"terminalBench40":0.0151515151515152,"enterpriseOpsGym":0.4500149209191286,"gdpvalNormalized":0.313825,"intelligenceIndex":29.4571634255623,"terminalbenchHard":0.507575757575758,"briefcaseTotalCost":118.41907193994186,"price1mInputTokens":2.5,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.31133333333333335,"price1mOutputTokens":7.5,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":5,"price1mBlended0To3To1":3.75,"price1mBlended7To2To1":1.6,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":2.5495049504950495,"price1mBlended100To1To1":0.5882352941176471,"automationBenchPartialScore":0.22976890336173603,"intelligenceIndexTimePerTask":137.2323991230729,"omniscienceHallucinationRate":0.2562923523717328,"inferenceParametersActiveBillions":null},"name":"Qwen3.7 Max","slug":"qwen3-7-max","source":"https://artificialanalysis.ai/models/qwen3-7-max","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.7-max/benchmarks"},"created":0,"released_at":"2026-05-21","release_source":"https://docs.qwencloud.com/changelog/models","description":"Qwen3.7 Max is the flagship of Alibaba's Qwen3.7 series, a text model designed for agent-centric workloads in coding, office productivity, and autonomous execution. It ships with thinking mode enabled by default and supports explicit caching for repeated prompts.","description_sources":[{"url":"https://openrouter.ai/qwen/qwen3.7-max","title":"OpenRouter model listing"},{"url":"https://docs.qwencloud.com/changelog/models","title":"Qwen model changelog"}],"display_name":"Qwen3.7 Max","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":2.5,"output_per_mtok_usd":7.5,"cached_input_per_mtok_usd":0.5}},{"id":"qwen/qwen3.7-plus","providers":["qwen"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":27.3,"artificial_analysis_coding_index":55.9,"artificial_analysis_agentic_index":17.5,"artificial_analysis_economics_index":37.3,"artificial_analysis_engineering_index":28.8,"artificial_analysis_intelligence_index":25.2,"artificial_analysis_strategy_and_ops_index":18.7,"artificial_analysis_finance_and_accounting_index":24.5,"artificial_analysis_healthcare_and_medical_index":23}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.356348470806302,"lcr":0.73,"gpqa":0.9,"tau2":0.929824561403509,"aime25":null,"critpt":0.0914285714285714,"gdpval":767.63,"suffix":null,"ifbench":0.779591836734694,"mmmuPro":0.804624277456647,"scicode":0.460648148148148,"apexAgents":0.224188790560472,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.175257731958763,"licenseName":null,"mlcrOverall":0.0888888888888889,"omniscience":1.0833333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.08,"gdpPdfAllPass":0.122,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.610486891385768,"terminalBench40":0.0101010101010101,"enterpriseOpsGym":0.40555058191584603,"gdpvalNormalized":0.133815,"intelligenceIndex":25.1622215821984,"terminalbenchHard":0.46969696969697,"briefcaseTotalCost":21.8308320957127,"price1mInputTokens":0.4,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.22516666666666665,"price1mOutputTokens":1.6,"pricePer1k1mpImages":0.4104,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":1,"price1mBlended0To3To1":0.7000000000000001,"price1mBlended7To2To1":0.296,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.4118811881188119,"price1mBlended100To1To1":0.09803921568627451,"automationBenchPartialScore":0.17391220980015495,"intelligenceIndexTimePerTask":580.0643309653647,"omniscienceHallucinationRate":0.2766186276618628,"inferenceParametersActiveBillions":null},"name":"Qwen3.7 Plus","slug":"qwen3-7-plus","source":"https://artificialanalysis.ai/models/qwen3-7-plus","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.7-plus/benchmarks"},"created":0,"released_at":"2026-06-01","release_source":"https://docs.qwencloud.com/changelog/models","description":"Qwen3.7 Plus is the cost-effective mid-tier model in Alibaba's Qwen3.7 series, accepting text and image input. It comprehensively upgrades vision-language abilities while keeping full agentic capabilities, targeting multimodal hybrid agent tasks such as screen reading, GUI operation, and mobile app navigation.","description_sources":[{"url":"https://openrouter.ai/qwen/qwen3.7-plus","title":"OpenRouter model listing"},{"url":"https://docs.qwencloud.com/changelog/models","title":"Qwen model changelog"}],"display_name":"Qwen3.7 Plus","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.4,"output_per_mtok_usd":1.6,"cached_input_per_mtok_usd":0.08,"context_price_tiers":[{"above_input_tokens":256000,"input_per_mtok_usd":1.2,"output_per_mtok_usd":4.8,"cached_input_per_mtok_usd":0.24}]}},{"id":"qwen/qwen3.6-flash","providers":["qwen"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:38:55.901Z","nextCheckAt":"2026-10-28T20:38:55.901Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.6-flash/benchmarks"},"created":0,"released_at":"2026-04-16","release_source":"https://docs.qwencloud.com/changelog/models","description":"Qwen3.6 Flash is the fast, efficient model in Alibaba's Qwen3.6 series, accepting text, image, and video input. It delivers broad improvements over Qwen3.5 Flash, with enhanced agent programming, math and code reasoning, and spatial intelligence, plus long-context inputs on a tiered billing schedule.","description_sources":[{"url":"https://openrouter.ai/qwen/qwen3.6-flash","title":"OpenRouter model listing"},{"url":"https://docs.qwencloud.com/changelog/models","title":"Qwen model changelog"}],"display_name":"Qwen3.6 Flash","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.25,"output_per_mtok_usd":1.5,"cached_input_per_mtok_usd":0.05,"context_price_tiers":[{"above_input_tokens":256000,"input_per_mtok_usd":1,"output_per_mtok_usd":4,"cached_input_per_mtok_usd":0.2}]}},{"id":"qwen/qwen3.6-27b","providers":["qwen","groq"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"groq","options":[{"region":null,"context_window":131072,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":23.3,"artificial_analysis_coding_index":53.7,"artificial_analysis_agentic_index":18.5,"artificial_analysis_economics_index":31.4,"artificial_analysis_engineering_index":23.3,"artificial_analysis_intelligence_index":21.4,"artificial_analysis_strategy_and_ops_index":16.9,"artificial_analysis_finance_and_accounting_index":21.3,"artificial_analysis_healthcare_and_medical_index":19.3}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-23T03:28:51.235Z"},"data":{"hle":0.230769230769231,"lcr":0.773333333333333,"gpqa":0.842424242424242,"tau2":0.941520467836257,"aime25":null,"critpt":0.0114285714285714,"gdpval":982.56,"ifbench":0.675510204081633,"mmmuPro":0.746242774566474,"scicode":0.428240740740741,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":27.8,"tauBanking":0.167010309278351,"mlcrOverall":0.105555555555556,"omniscience":-20.016666666666666,"analystAgent":null,"frontierCode":null,"cacheHitPrice":null,"gdpPdfAllPass":0.11,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":null,"reasoningTokens":5676,"terminalBench21":0.606741573033708,"terminalBench40":0,"gdpvalNormalized":0.24127999999999997,"intelligenceIndex":21.4279902240782,"terminalbenchHard":0.348484848484849,"briefcaseTotalCost":95.2408914,"price1mInputTokens":0.6,"contextWindowTokens":262144,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.196,"price1mOutputTokens":3.6,"pricePer1k1mpImages":0.615,"terminalBenchScience":null,"price1mBlended0To1To1":2.1,"price1mBlended0To3To1":1.35,"price1mBlended7To2To1":0.9,"cacheHitDiscountPercent":null,"price1mBlended0To100To1":0.6297029702970297,"price1mBlended100To1To1":0.6294117647058823,"automationBenchPartialScore":0.11045223588713587,"intelligenceIndexTimePerTask":446.6575075371082,"omniscienceHallucinationRate":0.49274461028192373,"inferenceParametersActiveBillions":null},"name":"Qwen3.6 27B (Reasoning)","slug":"qwen3-6-27b","source":"https://artificialanalysis.ai/models/qwen3-6-27b","fetchedAt":"2026-10-23T03:28:51.235Z","pageRecords":[]}],"checkedAt":"2026-10-23T03:28:51.235Z","fetchedAt":"2026-10-23T03:28:51.235Z","nextCheckAt":"2026-10-30T03:28:51.235Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.6-27b/benchmarks"},"created":0,"released_at":"2026-04-22","release_source":"https://qwen.ai/blog?id=qwen3.6-27b","description":"Qwen3.6 27B is the Qwen team's dense 27-billion-parameter open-weight model, released under Apache 2.0 and built on direct community feedback for stability and everyday utility. It targets agentic coding — repository-level comprehension and front-end workflows — and adds an option to preserve reasoning from earlier turns so iterative sessions keep their context.","description_sources":[{"url":"https://qwen.ai/blog?id=qwen3.6-27b","title":"Qwen blog: Qwen3.6-27B"},{"url":"https://huggingface.co/Qwen/Qwen3.6-27B","title":"Qwen/Qwen3.6-27B model card"},{"url":"https://openrouter.ai/qwen/qwen3.6-27b","title":"OpenRouter model listing"}],"display_name":"Qwen3.6 27B","context_window":229376,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.6,"output_per_mtok_usd":3.6}},{"id":"qwen/qwen3.8-max","providers":["qwen"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":46.1,"artificial_analysis_coding_index":76.2,"artificial_analysis_agentic_index":56,"artificial_analysis_economics_index":52.3,"artificial_analysis_engineering_index":46.4,"artificial_analysis_intelligence_index":45.4,"artificial_analysis_strategy_and_ops_index":47.3,"artificial_analysis_finance_and_accounting_index":45.6,"artificial_analysis_healthcare_and_medical_index":41.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.430954587581094,"lcr":0.803333333333333,"gpqa":0.928282828282828,"tau2":null,"aime25":null,"critpt":0.177142857142857,"gdpval":1664.63,"suffix":null,"aa3dElo":948.4151,"ifbench":null,"mmmuPro":0.827745664739884,"scicode":0.520833333333333,"apexAgents":0.424041297935103,"hleDiamond":null,"itBenchSre":0.402542372881356,"licenseUrl":null,"parameters":null,"tauBanking":0.478350515463918,"licenseName":null,"mlcrOverall":0.2,"omniscience":11.983333333333333,"analystAgent":0.45,"frontierCode":0.431563,"cacheHitPrice":0.25,"gdpPdfAllPass":0.228,"livecodebench":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.887640449438202,"terminalBench40":0.388888888888889,"enterpriseOpsGym":0.4762757385854969,"gdpvalNormalized":0.582315,"intelligenceIndex":45.4152084980521,"terminalbenchHard":null,"briefcaseTotalCost":1781.494041806632,"price1mInputTokens":2,"contextWindowTokens":983616,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.31683333333333336,"price1mOutputTokens":6,"pricePer1k1mpImages":null,"terminalBenchScience":0.119047619047619,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.175,"cacheHitDiscountPercent":0.875,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.3235294117647059,"automationBenchPartialScore":0.561905381087678,"intelligenceIndexTimePerTask":2063.244489880559,"omniscienceHallucinationRate":0.28836301536960235,"inferenceParametersActiveBillions":null},"name":"Qwen3.8 Max (0902)","slug":"qwen3-8-max","source":"https://artificialanalysis.ai/models/qwen3-8-max","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]},{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":41.5,"artificial_analysis_coding_index":71.8,"artificial_analysis_agentic_index":49.1,"artificial_analysis_economics_index":48.5,"artificial_analysis_engineering_index":40.8,"artificial_analysis_intelligence_index":40.2,"artificial_analysis_strategy_and_ops_index":42.2,"artificial_analysis_finance_and_accounting_index":40.3,"artificial_analysis_healthcare_and_medical_index":38.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.430491195551437,"lcr":0.783333333333333,"gpqa":0.927272727272727,"tau2":null,"aime25":null,"critpt":0.2,"gdpval":1604.43,"suffix":null,"ifbench":null,"mmmuPro":0.823121387283237,"scicode":0.532407407407407,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":0.51340206185567,"licenseName":null,"mlcrOverall":0.261111111111111,"omniscience":3.4,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.25,"gdpPdfAllPass":0.202,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.812734082397004,"terminalBench40":0.186868686868687,"gdpvalNormalized":0.552215,"intelligenceIndex":40.1528623585941,"terminalbenchHard":null,"briefcaseTotalCost":387.24121474095364,"price1mInputTokens":2,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.3185,"price1mOutputTokens":6,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.175,"cacheHitDiscountPercent":0.875,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.3235294117647059,"automationBenchPartialScore":0.4923210407553669,"intelligenceIndexTimePerTask":1223.5936775963605,"omniscienceHallucinationRate":0.417461482024945,"inferenceParametersActiveBillions":null},"name":"Qwen3.8 Max","slug":"qwen3-8-max-0803","source":"https://artificialanalysis.ai/models/qwen3-8-max-0803","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.8-max/benchmarks"},"created":0,"released_at":"2026-08-03","release_source":"https://www.alibabacloud.com/en/press-room/alibaba-unveils-qwen3-8-max","description":"Qwen3.8 Max is Alibaba's flagship Qwen model, a sparse mixture-of-experts vision-language model for complex reasoning, code generation, and long-document analysis. It can ingest hundred-page documents, full television series, or long livestreams, and has run multi-day autonomous software engineering projects.","description_sources":[{"url":"https://www.alibabacloud.com/en/press-room/alibaba-unveils-qwen3-8-max","title":"Alibaba Cloud press release"},{"url":"https://docs.qwencloud.com/changelog/models","title":"Qwen model changelog"}],"display_name":"Qwen3.8 Max","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":6,"cached_input_per_mtok_usd":0.25,"cache_write_per_mtok_usd":2.5}},{"id":"qwen/qwen3.8-27b","providers":["qwen","groq"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"groq","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":34.6,"artificial_analysis_coding_index":68.1,"artificial_analysis_agentic_index":45.8,"artificial_analysis_economics_index":42.8,"artificial_analysis_engineering_index":33.3,"artificial_analysis_intelligence_index":33.7,"artificial_analysis_strategy_and_ops_index":37.1,"artificial_analysis_finance_and_accounting_index":34.3,"artificial_analysis_healthcare_and_medical_index":34.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"modelCount":4,"intelligenceIndex":33.6762858057476},"name":"Qwen3.8 27B","slug":"qwen3-8-27b","source":"https://artificialanalysis.ai/models/qwen3-8-27b","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.8-27b/benchmarks"},"created":0,"released_at":"2026-08-14","release_source":"https://github.com/QwenLM/Qwen3.8/blob/main/README.md","description":"Qwen3.8 27B is an open-weight dense vision-language model from Alibaba's Qwen3.8 family, suited to coding, professional workflows, research, and long-running agent tasks. It builds on Qwen3.6 27B with gains in coding and office productivity, and offers flexible thinking control via reasoning effort and preserved thinking.","description_sources":[{"url":"https://openrouter.ai/qwen/qwen3.8-27b","title":"OpenRouter model listing"},{"url":"https://github.com/QwenLM/Qwen3.8/blob/main/README.md","title":"Qwen3.8 GitHub README"}],"display_name":"Qwen3.8 27B","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.5,"output_per_mtok_usd":3,"cached_input_per_mtok_usd":0.1}},{"id":"qwen/qwen3.8-2.4t-a95b","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":41.7,"artificial_analysis_coding_index":71.9,"artificial_analysis_agentic_index":50.1,"artificial_analysis_economics_index":49,"artificial_analysis_engineering_index":40.3,"artificial_analysis_intelligence_index":39.9,"artificial_analysis_strategy_and_ops_index":44,"artificial_analysis_finance_and_accounting_index":42.6,"artificial_analysis_healthcare_and_medical_index":36}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.424467099165894,"lcr":0.803333333333333,"gpqa":0.935353535353535,"tau2":null,"aime25":null,"critpt":0.2,"gdpval":1607.97,"suffix":null,"ifbench":null,"mmmuPro":null,"scicode":0.540509259259259,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":2400,"tauBanking":0.490721649484536,"mlcrOverall":0.305555555555556,"omniscience":4.316666666666666,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.25,"gdpPdfAllPass":0.156,"livecodebench":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.820224719101124,"terminalBench40":0.111111111111111,"enterpriseOpsGym":0.4735899731423456,"gdpvalNormalized":0.5539850000000001,"intelligenceIndex":39.9461723105325,"terminalbenchHard":null,"briefcaseTotalCost":378.1235622222578,"price1mInputTokens":2,"contextWindowTokens":262000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.3125,"price1mOutputTokens":6,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":4,"price1mBlended0To3To1":3,"price1mBlended7To2To1":1.175,"cacheHitDiscountPercent":0.875,"price1mBlended0To100To1":2.0396039603960396,"price1mBlended100To1To1":0.3235294117647059,"automationBenchPartialScore":0.572473765458222,"intelligenceIndexTimePerTask":1327.1617129483675,"omniscienceHallucinationRate":0.39175757575757575,"inferenceParametersActiveBillions":95},"name":"Qwen3.8 2.4T A95B","slug":"qwen3-8-2-4t-a95b","source":"https://artificialanalysis.ai/models/qwen3-8-2-4t-a95b","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.8-2.4t-a95b/benchmarks"},"created":0,"released_at":"2026-08-12","release_source":"https://github.com/QwenLM/Qwen3.8/blob/main/README.md","description":"Qwen3.8 2.4T A95B is the open-weight flagship of Alibaba's Qwen3.8 family, a sparse mixture-of-experts vision-language model and the open release of Qwen3.8 Max. It targets coding, productivity, research, and long-horizon agent tasks, with stronger autonomous planning and better handling of environment feedback.","description_sources":[{"url":"https://openrouter.ai/qwen/qwen3.8-2.4t-a95b","title":"OpenRouter model listing"},{"url":"https://github.com/QwenLM/Qwen3.8/blob/main/README.md","title":"Qwen3.8 GitHub README"}],"display_name":"Qwen3.8 2.4T A95B","context_window":262144,"reasoning":{"kind":"effort","levels":["low","medium","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":1.2,"output_per_mtok_usd":1.2,"cached_input_per_mtok_usd":0.6}},{"id":"xiaomi/mimo-v2.5","providers":["atlascloud"],"provider_details":[{"provider":"atlascloud","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":24.4,"artificial_analysis_coding_index":56.8,"artificial_analysis_agentic_index":15.8,"artificial_analysis_economics_index":33.9,"artificial_analysis_engineering_index":24.6,"artificial_analysis_intelligence_index":21.8,"artificial_analysis_strategy_and_ops_index":18.9,"artificial_analysis_finance_and_accounting_index":22.2,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.272011121408712,"lcr":0.73,"gpqa":0.84949494949495,"tau2":0.906432748538012,"aime25":null,"critpt":0.0371428571428571,"gdpval":997.19,"suffix":null,"ifbench":0.671428571428571,"mmmuPro":0.754335260115607,"scicode":0.438657407407407,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":310,"tauBanking":0.0865979381443299,"omniscience":-9.833333333333334,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.0028,"gdpPdfAllPass":0.04,"livecodebench":null,"hostModelCount":5,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.636704119850187,"terminalBench40":0,"gdpvalNormalized":0.24859500000000004,"intelligenceIndex":21.8248678794953,"terminalbenchHard":0.416666666666667,"briefcaseTotalCost":1.415279430252699,"price1mInputTokens":0.14,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.1675,"price1mOutputTokens":0.28,"pricePer1k1mpImages":1.026,"terminalBenchScience":null,"price1mBlended0To1To1":0.21000000000000002,"price1mBlended0To3To1":0.17500000000000002,"price1mBlended7To2To1":0.05796000000000001,"cacheHitDiscountPercent":0.98,"price1mBlended0To100To1":0.1413861386138614,"price1mBlended100To1To1":0.006862745098039216,"automationBenchPartialScore":0.1842951233004315,"intelligenceIndexTimePerTask":884.1193427536706,"omniscienceHallucinationRate":0.31931931931931934,"inferenceParametersActiveBillions":15},"name":"MiMo-V2.5","slug":"mimo-v2-5-0424","source":"https://artificialanalysis.ai/models/mimo-v2-5-0424","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/xiaomi/mimo-v2.5/benchmarks"},"created":0,"released_at":"2026-04-23","release_source":"https://mimo.mi.com/docs/en-US/updates/model","description":"MiMo V2.5 is Xiaomi's native omnimodal model, perceiving images, video, audio, and text with native agent execution for tasks such as browsing. It sits on Xiaomi's performance-efficiency frontier, matching MiMo V2.5 Pro on daily tasks while surpassing MiMo V2 Omni in multimodal perception and long-range reasoning.","description_sources":[{"url":"https://openrouter.ai/xiaomi/mimo-v2.5","title":"OpenRouter model listing"},{"url":"https://mimo.mi.com/docs/en-US/updates/model","title":"Xiaomi MiMo model updates"}],"display_name":"MiMo V2.5","context_window":1024000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.14,"output_per_mtok_usd":0.28,"cached_input_per_mtok_usd":0.003}},{"id":"minimaxai/minimax-m3.1-flash","providers":["minimax","fireworks"],"provider_details":[{"provider":"minimax","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["AA API HTTP 429"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":46.8,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":50,"artificial_analysis_engineering_index":43.4,"artificial_analysis_intelligence_index":42.4,"artificial_analysis_strategy_and_ops_index":54.6,"artificial_analysis_finance_and_accounting_index":49,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-19T17:19:52.584Z"},"data":{"hle":0.479610750695088,"lcr":0.796666666666667,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.222857142857143,"gdpval":1465.71,"suffix":null,"ifbench":null,"mmmuPro":0.791329479768786,"scicode":0.533564814814815,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"healthbench":null,"licenseName":null,"mlcrOverall":0.166666666666667,"omniscience":15.55,"analystAgent":null,"frontierCode":0.385759,"cacheHitPrice":0.015,"gdpPdfAllPass":0.208,"livecodebench":null,"frontierCode20":null,"hostModelCount":15,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.257575757575758,"critptAgenticHard":null,"intelligenceIndex":42.7057475126235,"terminalbenchHard":null,"briefcaseTotalCost":48.103196798033665,"price1mInputTokens":0.15,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.562,"price1mOutputTokens":0.6,"pricePer1k1mpImages":null,"terminalBenchScience":0.0333333333333333,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":0.375,"price1mBlended0To3To1":0.26249999999999996,"price1mBlended7To2To1":0.10049999999999999,"cacheHitDiscountPercent":0.9,"healthbenchProfessional":null,"price1mBlended0To100To1":0.15445544554455445,"price1mBlended100To1To1":0.022058823529411766,"automationBenchPartialScore":0.7252180957532544,"intelligenceIndexTimePerTask":582.8126815475363,"omniscienceHallucinationRate":0.928082191780822,"inferenceParametersActiveBillions":null},"name":"MiniMax-M3.1-Flash","slug":"minimax-m3-1-flash","source":"https://artificialanalysis.ai/models/minimax-m3-1-flash","fetchedAt":"2026-10-27T18:03:56.749Z","pageRecords":[]}],"checkedAt":"2026-10-27T18:03:56.749Z","fetchedAt":"2026-10-19T17:19:52.584Z","nextCheckAt":"2026-10-28T18:03:56.749Z","summary":true,"detailPath":"/v1/models/minimaxai/minimax-m3.1-flash/benchmarks"},"created":0,"released_at":"2026-10-19","release_source":"https://www.minimax.io/blog/minimax-m31f","description":"MiniMax M3.1 Flash is MiniMax's fast, low-cost successor to M3, built for agentic work, coding, computer use and long-context tasks across a 1M-token window. Its reasoning is always on and tunable across five effort levels, and it accepts text and image input with tool calling.","description_sources":[{"url":"https://www.minimax.io/blog/minimax-m31f","title":"MiniMax release announcement"},{"url":"https://platform.minimax.io/docs/guides/pricing","title":"MiniMax pricing"}],"display_name":"MiniMax M3.1 Flash","context_window":1000000,"reasoning":{"kind":"effort","levels":["low","medium","high","xhigh","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.15,"output_per_mtok_usd":0.6,"cached_input_per_mtok_usd":0.015}},{"id":"minimaxai/minimax-m3","providers":["azure-foundry","minimax","fireworks"],"provider_details":[{"provider":"azure-foundry","options":[{"region":"centralus","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus2","context_window":null,"supports_tools":null,"route_count":1},{"region":"northcentralus","context_window":null,"supports_tools":null,"route_count":1},{"region":"eastus","context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"minimax","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":31.5,"artificial_analysis_coding_index":58.6,"artificial_analysis_agentic_index":29.5,"artificial_analysis_economics_index":43.7,"artificial_analysis_engineering_index":31.2,"artificial_analysis_intelligence_index":29.2,"artificial_analysis_strategy_and_ops_index":25,"artificial_analysis_finance_and_accounting_index":30.1,"artificial_analysis_healthcare_and_medical_index":29.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.389712696941613,"lcr":0.83,"gpqa":0.929292929292929,"tau2":0.888888888888889,"aime25":null,"critpt":0.0371428571428571,"gdpval":1242.47,"suffix":null,"aa3dElo":740.2086,"ifbench":0.828571428571429,"mmmuPro":0.785549132947977,"scicode":0.471064814814815,"harveyLab":0,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":428,"tauBanking":0.152577319587629,"mlcrOverall":0.0444444444444444,"omniscience":1.35,"analystAgent":0.1,"frontierCode":0.230037666666667,"cacheHitPrice":0.06,"gdpPdfAllPass":0.098,"livecodebench":null,"hostModelCount":14,"cacheWritePrice":0.375,"reasoningTokens":2000,"terminalBench21":0.651685393258427,"terminalBench40":0.0202020202020202,"enterpriseOpsGym":0.3210981796478663,"gdpvalNormalized":0.37123500000000004,"intelligenceIndex":29.2202932236316,"terminalbenchHard":0.424242424242424,"briefcaseTotalCost":85.49718027093401,"price1mInputTokens":0.3,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.167,"price1mOutputTokens":1.2,"pricePer1k1mpImages":null,"terminalBenchScience":0.00476190476190476,"price1mBlended0To1To1":0.75,"price1mBlended0To3To1":0.5249999999999999,"price1mBlended7To2To1":0.22199999999999998,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.3089108910891089,"price1mBlended100To1To1":0.07352941176470588,"automationBenchPartialScore":0.21251257600749407,"intelligenceIndexTimePerTask":568.6357703403531,"omniscienceHallucinationRate":0.1842737094837935,"inferenceParametersActiveBillions":23},"name":"MiniMax-M3","slug":"minimax-m3","source":"https://artificialanalysis.ai/models/minimax-m3","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/minimaxai/minimax-m3/benchmarks"},"created":0,"released_at":"2026-06-01","release_source":"https://www.minimax.io/blog/minimax-m3","description":"MiniMax M3 is MiniMax's frontier open-weight coding model, combining frontier-level coding and agentic performance with ultra-long context and native multimodality. It introduces MiniMax Sparse Attention for fast long-context prefill, accepts image and video input, and can operate a desktop computer.","description_sources":[{"url":"https://www.minimax.io/blog/minimax-m3","title":"MiniMax release announcement"},{"url":"https://platform.minimax.io/docs/guides/models-intro","title":"MiniMax model documentation"}],"display_name":"MiniMax M3","context_window":524300,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.3,"output_per_mtok_usd":1.2,"cached_input_per_mtok_usd":0.06,"context_price_tiers":[{"above_input_tokens":512000,"input_per_mtok_usd":0.6,"output_per_mtok_usd":2.4,"cached_input_per_mtok_usd":0.12}]}},{"id":"tencent/hy3","providers":["tencent","atlascloud"],"provider_details":[{"provider":"tencent","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"atlascloud","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":26.5,"artificial_analysis_coding_index":58.8,"artificial_analysis_agentic_index":24.1,"artificial_analysis_economics_index":35.2,"artificial_analysis_engineering_index":26.7,"artificial_analysis_intelligence_index":25.3,"artificial_analysis_strategy_and_ops_index":23.9,"artificial_analysis_finance_and_accounting_index":27.3,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.334569045412419,"lcr":0.79,"gpqa":0.896969696969697,"tau2":null,"aime25":null,"critpt":0.0485714285714286,"gdpval":1066.59,"suffix":null,"ifbench":null,"mmmuPro":null,"scicode":0.486111111111111,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":299,"tauBanking":0.228865979381443,"omniscience":-18.466666666666665,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.034,"gdpPdfAllPass":0.112,"livecodebench":null,"hostModelCount":4,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.644194756554307,"terminalBench40":0.00505050505050505,"gdpvalNormalized":0.28329499999999996,"intelligenceIndex":25.2972775731571,"terminalbenchHard":null,"briefcaseTotalCost":10.318991407699059,"price1mInputTokens":0.136,"contextWindowTokens":256000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.31966666666666665,"price1mOutputTokens":0.5549999999999999,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.3455,"price1mBlended0To3To1":0.24075,"price1mBlended7To2To1":0.1065,"cacheHitDiscountPercent":0.75,"price1mBlended0To100To1":0.14014851485148516,"price1mBlended100To1To1":0.040107843137254906,"automationBenchPartialScore":0.18409128371172395,"intelligenceIndexTimePerTask":536.5591000149514,"omniscienceHallucinationRate":0.7413032827045566,"inferenceParametersActiveBillions":21},"name":"Hy3","slug":"hy3","source":"https://artificialanalysis.ai/models/hy3","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/tencent/hy3/benchmarks"},"created":0,"released_at":"2026-07-06","release_source":"https://www.tencent.com/tencent-hunyuan-officially-releases-hy3-advancing-agent-capabilities-and-deeper-product-integration/","description":"Hunyuan 3 is Tencent's open-weight mixture-of-experts model, released under Apache 2.0 for software development, office productivity, and financial modeling. Over Hy2 it improves complex reasoning, instruction following, code generation, and agent capabilities, and exposes configurable reasoning effort.","description_sources":[{"url":"https://openrouter.ai/tencent/hy3","title":"OpenRouter model listing"},{"url":"https://www.tencent.com/tencent-hunyuan-officially-releases-hy3-advancing-agent-capabilities-and-deeper-product-integration/","title":"Tencent release announcement"}],"display_name":"Hunyuan 3","context_window":262144,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.132,"output_per_mtok_usd":0.528,"cached_input_per_mtok_usd":0.033}},{"id":"tencent/hy4-preview","providers":["tencent"],"provider_details":[{"provider":"tencent","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:38:55.901Z","nextCheckAt":"2026-10-28T20:38:55.901Z","summary":true,"detailPath":"/v1/models/tencent/hy4-preview/benchmarks"},"created":0,"description":"Hunyuan 4 Preview is Tencent's 770B-parameter mixture-of-experts model with 49B active parameters, served natively on Tencent Cloud TokenHub with a one-million-token window of up to 960K input and 64K output. It reasons at high effort by default and exposes a two-rung reasoning control, none or high, for callers who want to turn thinking off.","description_sources":[{"url":"https://www.tencentcloud.com/document/product/1300/78937","title":"Tencent Cloud TokenHub pricing"},{"url":"https://www.tencentcloud.com/document/product/1300/80637","title":"Tencent Cloud deep reasoning guide"}],"display_name":"Hunyuan 4 Preview","context_window":1048576,"reasoning":{"kind":"effort","levels":["high"],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.834,"output_per_mtok_usd":2.501,"cached_input_per_mtok_usd":0.042}},{"id":"thinkingmachines/inkling","providers":["thinkingmachines"],"provider_details":[{"provider":"thinkingmachines","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":31.4,"artificial_analysis_coding_index":52.1,"artificial_analysis_agentic_index":22.5,"artificial_analysis_economics_index":38.5,"artificial_analysis_engineering_index":27.7,"artificial_analysis_intelligence_index":25,"artificial_analysis_strategy_and_ops_index":23.2,"artificial_analysis_finance_and_accounting_index":29.5,"artificial_analysis_healthcare_and_medical_index":25.4}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.318813716404078,"lcr":0.773333333333333,"gpqa":0.871717171717172,"tau2":null,"aime25":null,"critpt":0.0542857142857143,"gdpval":1069.57,"aa3dElo":580.5701,"ifbench":null,"mmmuPro":0.734682080924856,"scicode":0.469907407407407,"harveyLab":0,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":975,"tauBanking":0.290721649484536,"mlcrOverall":0.0611111111111111,"omniscience":2,"analystAgent":0.2375,"frontierCode":0.158388333333333,"cacheHitPrice":0.17,"gdpPdfAllPass":0.128,"livecodebench":null,"hostModelCount":7,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.550561797752809,"terminalBench40":0.0101010101010101,"enterpriseOpsGym":0.37958818263205013,"gdpvalNormalized":0.28478499999999995,"intelligenceIndex":24.9847810999384,"terminalbenchHard":null,"price1mInputTokens":1,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.4155,"price1mOutputTokens":4.05,"pricePer1k1mpImages":null,"terminalBenchScience":0,"price1mBlended0To1To1":2.525,"price1mBlended0To3To1":1.7625,"price1mBlended7To2To1":0.724,"cacheHitDiscountPercent":0.83,"price1mBlended0To100To1":1.0301980198019802,"price1mBlended100To1To1":0.2161764705882353,"automationBenchPartialScore":0.04984889684475201,"intelligenceIndexTimePerTask":134.70625140458677,"omniscienceHallucinationRate":0.6766467065868264,"inferenceParametersActiveBillions":41},"name":"Inkling (Xhigh)","slug":"inkling","source":"https://artificialanalysis.ai/models/inkling","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/thinkingmachines/inkling/benchmarks"},"created":0,"released_at":"2026-07-15","release_source":"https://thinkingmachines.ai/news/introducing-inkling/","description":"Inkling is Thinking Machines Lab's open-weight multimodal mixture-of-experts model, a broad foundation model for reasoning, coding, agentic tool use, and fine-tuning. It processes text, images, audio, and video natively with an encoder-free design and offers controllable thinking effort to trade tokens for performance.","description_sources":[{"url":"https://openrouter.ai/thinkingmachines/inkling","title":"OpenRouter model listing"},{"url":"https://thinkingmachines.ai/news/introducing-inkling/","title":"Thinking Machines release announcement"}],"display_name":"Inkling","context_window":65536,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.87,"output_per_mtok_usd":4.68,"cached_input_per_mtok_usd":0.374}},{"id":"thinkingmachines/inkling-256k","providers":["thinkingmachines"],"provider_details":[{"provider":"thinkingmachines","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:38:55.901Z","nextCheckAt":"2026-10-28T20:38:55.901Z","summary":true,"detailPath":"/v1/models/thinkingmachines/inkling-256k/benchmarks"},"created":0,"released_at":"2026-07-15","release_source":"https://thinkingmachines.ai/news/introducing-inkling/","description":"Inkling 256K is the extended-context serving option for Thinking Machines Lab's Inkling, its open-weight multimodal mixture-of-experts model. It runs the same model with a larger context window than the default deployment, for long documents and extended agent sessions.","description_sources":[{"url":"https://thinkingmachines.ai/news/introducing-inkling/","title":"Thinking Machines release announcement"},{"url":"https://thinkingmachines.ai/inkling/","title":"Thinking Machines Inkling model page"}],"display_name":"Inkling 256K","context_window":262144,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":3.74,"output_per_mtok_usd":9.36,"cached_input_per_mtok_usd":0.748}},{"id":"meta/muse-wonder","providers":["meta"],"provider_details":[{"provider":"meta","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":52,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":57.7,"artificial_analysis_engineering_index":53.8,"artificial_analysis_intelligence_index":52.4,"artificial_analysis_strategy_and_ops_index":54.8,"artificial_analysis_finance_and_accounting_index":51.3,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"hle":0.500926784059314,"lcr":0.806666666666667,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.282857142857143,"gdpval":1770.34,"ifbench":null,"mmmuPro":null,"scicode":0.603009259259259,"harveyLab":0.1,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":null,"tauBanking":null,"licenseName":null,"omniscience":25.483333333333334,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.2,"gdpPdfAllPass":0.306,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.52020202020202,"gdpvalNormalized":0.63517,"intelligenceIndex":52.393894697293,"terminalbenchHard":null,"briefcaseTotalCost":1133.0033520700003,"price1mInputTokens":4,"contextWindowTokens":1048576,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.411,"price1mOutputTokens":20,"pricePer1k1mpImages":null,"terminalBenchScience":null,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":12,"price1mBlended0To3To1":8,"price1mBlended7To2To1":2.94,"cacheHitDiscountPercent":0.95,"price1mBlended0To100To1":4.158415841584159,"price1mBlended100To1To1":0.43137254901960786,"automationBenchPartialScore":0.6809455131207758,"intelligenceIndexTimePerTask":356.81547470627385,"omniscienceHallucinationRate":0.2651386530843237,"inferenceParametersActiveBillions":null},"name":"Muse Wonder (Max)","slug":"muse-wonder","source":"https://artificialanalysis.ai/models/muse-wonder","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/meta/muse-wonder/benchmarks"},"created":0,"released_at":"2026-10-12","release_source":"https://about.fb.com/news/2026/10/introducing-muse-wonder-models/","description":"Muse Wonder is Meta's reasoning model for complex software engineering and long-running personal and professional agents. It is built to track context across many steps, make and verify changes across a codebase, call tools in parallel, and recover from errors while following detailed instructions.","description_sources":[{"url":"https://dev.meta.ai/docs/models#muse-wonder","title":"Meta Model API: Muse Wonder"},{"url":"https://research.meta.ai/blog/introducing-muse-wonder/","title":"Meta research announcement: Muse Wonder"}],"display_name":"Muse Wonder","context_window":1048576,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high","xhigh","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":4,"output_per_mtok_usd":20,"cached_input_per_mtok_usd":0.2}},{"id":"meta/muse-spark-open","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":44.5,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":51,"artificial_analysis_engineering_index":40.4,"artificial_analysis_intelligence_index":38.9,"artificial_analysis_strategy_and_ops_index":42.7,"artificial_analysis_finance_and_accounting_index":42.5,"artificial_analysis_healthcare_and_medical_index":39}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-26T03:48:53.038Z"},"data":{"hle":0.438832252085264,"lcr":0.783333333333333,"gpqa":0.88989898989899,"tau2":null,"aime25":null,"critpt":0.148571428571429,"gdpval":1401.12,"ifbench":null,"mmmuPro":0.809248554913295,"scicode":0.584490740740741,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":816,"tauBanking":null,"mlcrOverall":0.0777777777777778,"omniscience":24.65,"analystAgent":null,"frontierCode":0.276476,"cacheHitPrice":0.15,"gdpPdfAllPass":0.186,"livecodebench":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.764044943820225,"terminalBench40":0.0808080808080808,"intelligenceIndex":38.8666662997805,"terminalbenchHard":null,"briefcaseTotalCost":133.35160182000004,"price1mInputTokens":1.25,"contextWindowTokens":1048576,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.39266666666666666,"price1mOutputTokens":4.25,"pricePer1k1mpImages":null,"terminalBenchScience":0.019047619047619,"price1mBlended0To1To1":2.75,"price1mBlended0To3To1":2,"price1mBlended7To2To1":0.78,"cacheHitDiscountPercent":0.88,"price1mBlended0To100To1":1.2797029702970297,"price1mBlended100To1To1":0.20098039215686275,"automationBenchPartialScore":0.4970811827313194,"intelligenceIndexTimePerTask":283.87551719214133,"omniscienceHallucinationRate":0.2406695938529089,"inferenceParametersActiveBillions":42},"name":"Muse Spark Open (Xhigh)","slug":"muse-spark-open","source":"https://artificialanalysis.ai/models/muse-spark-open","fetchedAt":"2026-10-26T03:48:53.038Z","pageRecords":[]}],"checkedAt":"2026-10-26T03:48:53.038Z","fetchedAt":"2026-10-26T03:48:53.038Z","nextCheckAt":"2026-11-02T03:48:53.038Z","summary":true,"detailPath":"/v1/models/meta/muse-spark-open/benchmarks"},"created":0,"released_at":"2026-10-15","release_source":"https://research.meta.ai/blog/muse-spark-open","description":"Muse Spark Open is Meta's open-weight multimodal reasoning model, a sparse mixture-of-experts with 816 billion total and 42 billion active parameters. It is the downloadable sibling of the hosted Muse Spark line, aimed at coding and long-horizon agentic work that you can run on your own hardware or any inference host.","description_sources":[{"url":"https://research.meta.ai/blog/muse-spark-open","title":"Meta AI Research: Introducing Muse Spark Open"},{"url":"https://openrouter.ai/meta/muse-spark-open","title":"OpenRouter: Muse Spark Open"},{"url":"https://api.fireworks.ai/v1/accounts/fireworks/models/muse-spark-open","title":"Fireworks AI model API: muse-spark-open"}],"display_name":"Muse Spark Open","context_window":1048576,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high","xhigh","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":4.25,"cached_input_per_mtok_usd":0.15}},{"id":"meta/muse-spark-1.3","providers":["meta"],"provider_details":[{"provider":"meta","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":50.9,"artificial_analysis_coding_index":75.8,"artificial_analysis_agentic_index":55.5,"artificial_analysis_economics_index":56.3,"artificial_analysis_engineering_index":49.4,"artificial_analysis_intelligence_index":48.1,"artificial_analysis_strategy_and_ops_index":51.2,"artificial_analysis_finance_and_accounting_index":49.8,"artificial_analysis_healthcare_and_medical_index":49.7}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:22:53.225Z"},"data":{"modelCount":2,"intelligenceIndex":48.0923107719685},"name":"Muse Spark 1.3","slug":"muse-spark-1-3","source":"https://artificialanalysis.ai/models/muse-spark-1-3","fetchedAt":"2026-10-21T17:22:53.225Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:22:53.225Z","fetchedAt":"2026-10-21T17:22:53.225Z","nextCheckAt":"2026-10-28T17:22:53.225Z","summary":true,"detailPath":"/v1/models/meta/muse-spark-1.3/benchmarks"},"created":0,"released_at":"2026-09-02","release_source":"https://research.meta.ai/blog/introducing-muse-spark-1-3","description":"Muse Spark 1.3 is Meta's multimodal reasoning model for long-running agentic, multi-agent, and coding workflows, taking text, image, and PDF input. Compared with Muse Spark 1.2 it takes fewer turns and is less verbose, uses fewer tool calls, and adds stronger resistance to adversarial inputs and prompt injections.","description_sources":[{"url":"https://openrouter.ai/meta/muse-spark-1.3","title":"OpenRouter model listing"},{"url":"https://research.meta.ai/blog/introducing-muse-spark-1-3","title":"Meta research announcement"}],"display_name":"Muse Spark 1.3","context_window":1048576,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high","xhigh","max"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.25,"output_per_mtok_usd":4.25,"cached_input_per_mtok_usd":0.15}},{"id":"meta/muse-spark-1.3-contributor","providers":["meta"],"provider_details":[{"provider":"meta","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:38:55.901Z","nextCheckAt":"2026-10-28T20:38:55.901Z","summary":true,"detailPath":"/v1/models/meta/muse-spark-1.3-contributor/benchmarks"},"created":0,"released_at":"2026-09-02","release_source":"https://vercel.com/changelog/muse-spark-1-3-now-available-on-ai-gateway","description":"Muse Spark 1.3 (Contributor) is the contributor billing tier of Meta's Muse Spark 1.3, the same multimodal reasoning model with identical weights and capabilities. Meta may use inputs and outputs sent to this tier to train its models, and pricing is lower in exchange.","description_sources":[{"url":"https://openrouter.ai/meta/muse-spark-1.3-contributor","title":"OpenRouter model listing"},{"url":"https://vercel.com/changelog/muse-spark-1-3-now-available-on-ai-gateway","title":"Vercel AI Gateway changelog"}],"display_name":"Muse Spark 1.3 (Contributor)","context_window":1048576,"reasoning":{"kind":"effort","levels":["minimal","low","medium","high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0.2,"cached_input_per_mtok_usd":0.002}},{"id":"qwen/qwen3.8-flash","providers":["qwen"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T20:38:55.901Z","nextCheckAt":"2026-10-28T20:38:55.901Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.8-flash/benchmarks"},"created":0,"description":"Qwen3.8 Flash is Alibaba's multimodal reasoning model for coding assistance, tool use, and visual understanding. It works with text and images to analyze documents, code, and charts while offering a switch for reasoning.","description_sources":[{"url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash","title":"Creator model documentation"},{"url":"https://openrouter.ai/qwen/qwen3.8-flash","title":"OpenRouter model listing"}],"display_name":"Qwen3.8 Flash","context_window":1000000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.15,"output_per_mtok_usd":0.47,"cached_input_per_mtok_usd":0.016,"cache_write_per_mtok_usd":0.2}},{"id":"qwen/qwen3.7-flash","providers":["qwen"],"provider_details":[{"provider":"qwen","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/qwen/qwen3.7-flash/benchmarks"},"created":0,"description":"Qwen3.7 Flash is Alibaba's vision-language model for assistants that combine reasoning with image understanding. Its focus includes object recognition, spatial interpretation, and coding tasks that involve visual information.","description_sources":[{"url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-7-flash","title":"Creator model documentation"},{"url":"https://openrouter.ai/qwen/qwen3.7-flash","title":"OpenRouter model listing"}],"display_name":"Qwen3.7 Flash","context_window":1000000,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.03,"output_per_mtok_usd":0.13,"cached_input_per_mtok_usd":0.006,"cache_write_per_mtok_usd":0.038,"context_price_tiers":[{"above_input_tokens":32000,"input_per_mtok_usd":0.1,"output_per_mtok_usd":0.4,"cached_input_per_mtok_usd":0.02,"cache_write_per_mtok_usd":0.125},{"above_input_tokens":256000,"input_per_mtok_usd":0.2,"output_per_mtok_usd":0.8,"cached_input_per_mtok_usd":0.04,"cache_write_per_mtok_usd":0.25}]}},{"id":"thinkingmachines/inkling-small","providers":["thinkingmachines"],"provider_details":[{"provider":"thinkingmachines","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":28.2,"artificial_analysis_coding_index":52.9,"artificial_analysis_agentic_index":23.5,"artificial_analysis_economics_index":37.7,"artificial_analysis_engineering_index":28.9,"artificial_analysis_intelligence_index":25.7,"artificial_analysis_strategy_and_ops_index":21.3,"artificial_analysis_finance_and_accounting_index":27.7,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"hle":0.333178869323448,"lcr":0.756666666666667,"gpqa":0.894949494949495,"tau2":null,"aime25":null,"critpt":0.0828571428571429,"gdpval":1122.78,"suffix":null,"ifbench":null,"mmmuPro":0.740462427745665,"scicode":0.496527777777778,"harveyLab":0,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":266,"tauBanking":0.187628865979381,"mlcrOverall":0.0777777777777778,"omniscience":-8.933333333333334,"analystAgent":0.275,"frontierCode":0.00553333333333333,"cacheHitPrice":0.06,"gdpPdfAllPass":0.112,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.550561797752809,"terminalBench40":0.0101010101010101,"enterpriseOpsGym":0.4270367054610564,"gdpvalNormalized":0.31139,"intelligenceIndex":25.6582584411942,"terminalbenchHard":null,"briefcaseTotalCost":10.741384859182709,"price1mInputTokens":0.3,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.33166666666666667,"price1mOutputTokens":1.2,"pricePer1k1mpImages":null,"terminalBenchScience":0,"price1mBlended0To1To1":0.75,"price1mBlended0To3To1":0.5249999999999999,"price1mBlended7To2To1":0.22199999999999998,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.3089108910891089,"price1mBlended100To1To1":0.07352941176470588,"automationBenchPartialScore":0.048142275152503786,"intelligenceIndexTimePerTask":131.93616950230603,"omniscienceHallucinationRate":0.629925187032419,"inferenceParametersActiveBillions":12},"name":"Inkling Small","slug":"inkling-small","source":"https://artificialanalysis.ai/models/inkling-small","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/thinkingmachines/inkling-small/benchmarks"},"created":0,"description":"Inkling Small is Thinking Machines Lab's smaller open-weight generalist for reasoning, coding, and tool use. Its text and image understanding supports work with documents, charts, and diagrams.","description_sources":[{"url":"https://thinkingmachines.ai/news/inkling-small/","title":"Thinking Machines release announcement"},{"url":"https://tinker-docs.thinkingmachines.ai/tinker/models/","title":"Thinking Machines model documentation"}],"display_name":"Inkling Small","context_window":65536,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.58,"output_per_mtok_usd":1.44,"cached_input_per_mtok_usd":0.116}},{"id":"thinkingmachines/inkling-small-256k","providers":["thinkingmachines"],"provider_details":[{"provider":"thinkingmachines","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/thinkingmachines/inkling-small-256k/benchmarks"},"created":0,"description":"Inkling Small 256K is the extended-context serving option for Thinking Machines Lab's smaller open-weight generalist. It combines text and image understanding for longer documents, coding tasks, and conversations.","description_sources":[{"url":"https://tinker-docs.thinkingmachines.ai/tinker/models/","title":"Thinking Machines model documentation"},{"url":"https://thinkingmachines.ai/news/inkling-small/","title":"Thinking Machines release announcement"}],"display_name":"Inkling Small 256K","context_window":262144,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.16,"output_per_mtok_usd":2.89,"cached_input_per_mtok_usd":0.232}},{"id":"nvidia/nemotron-3-ultra-550b-a55b","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["nemotron-3-ultra-550b-a55b: AA page HTTP 404"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":26.4,"artificial_analysis_coding_index":49.3,"artificial_analysis_agentic_index":20.1,"artificial_analysis_economics_index":36.6,"artificial_analysis_engineering_index":27.2,"artificial_analysis_intelligence_index":22.9,"artificial_analysis_strategy_and_ops_index":16.4,"artificial_analysis_finance_and_accounting_index":23.9,"artificial_analysis_healthcare_and_medical_index":23}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-14T17:26:48.816Z"},"data":{"hle":0.284059314179796,"lcr":0.793333333333333,"gpqa":0.866666666666667,"tau2":0.833333333333333,"aime25":null,"critpt":0.0314285714285714,"gdpval":1013.25,"ifbench":0.813605442176871,"mmmuPro":null,"scicode":0.402777777777778,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":550,"tauBanking":0.142268041237113,"mlcrOverall":0.111111111111111,"omniscience":-0.4,"analystAgent":0.0625,"frontierCode":0.105152333333333,"cacheHitPrice":0.16,"gdpPdfAllPass":0.05,"livecodebench":null,"hostModelCount":8,"cacheWritePrice":null,"reasoningTokens":2275,"terminalBench21":0.539325842696629,"terminalBench40":0.00505050505050505,"enterpriseOpsGym":0.28886899433005075,"gdpvalNormalized":0.256625,"intelligenceIndex":22.9279123422858,"terminalbenchHard":0.363636363636364,"briefcaseTotalCost":121.23588845964517,"price1mInputTokens":0.6,"contextWindowTokens":262144,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.22583333333333333,"price1mOutputTokens":2.5,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":1.55,"price1mBlended0To3To1":1.075,"price1mBlended7To2To1":0.48200000000000004,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.6188118811881188,"price1mBlended100To1To1":0.18725490196078431,"automationBenchPartialScore":0.02986349541593992,"intelligenceIndexTimePerTask":165.77309937960604,"omniscienceHallucinationRate":0.2968783638320775,"inferenceParametersActiveBillions":55},"name":"Nemotron 3 Ultra 550B A55B (Reasoning)","slug":"nvidia-nemotron-3-ultra-550b-a55b","source":"https://artificialanalysis.ai/models/nvidia-nemotron-3-ultra-550b-a55b","fetchedAt":"2026-10-14T17:26:48.816Z","pageRecords":[]}],"checkedAt":"2026-10-27T21:09:52.480Z","fetchedAt":"2026-10-14T17:26:48.816Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/nvidia/nemotron-3-ultra-550b-a55b/benchmarks"},"created":0,"description":"Nemotron 3 Ultra is NVIDIA's open-weight language model for reasoning, coding, and coordinating tasks through tools. Its hybrid architecture combines attention with state-space and mixture-of-experts layers, and applications can turn reasoning on or off.","description_sources":[{"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4","title":"NVIDIA model card"},{"url":"https://fireworks.ai/models/fireworks/nemotron-3-ultra-nvfp4","title":"Fireworks model documentation"}],"display_name":"Nemotron 3 Ultra 550B A55B","context_window":262144,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.6,"output_per_mtok_usd":2.4,"cached_input_per_mtok_usd":0.12}},{"id":"moonshotai/kimi-k2.7-code","providers":["moonshotai"],"provider_details":[{"provider":"moonshotai","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":27.7,"artificial_analysis_coding_index":60.8,"artificial_analysis_agentic_index":21,"artificial_analysis_economics_index":36.5,"artificial_analysis_engineering_index":29.1,"artificial_analysis_intelligence_index":25.8,"artificial_analysis_strategy_and_ops_index":28,"artificial_analysis_finance_and_accounting_index":28.9,"artificial_analysis_healthcare_and_medical_index":26.5}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"hle":0.35032437442076,"lcr":0.793333333333333,"gpqa":0.895959595959596,"tau2":0.900584795321637,"aime25":null,"critpt":0.1,"gdpval":1037.51,"suffix":null,"ifbench":0.631292517006803,"mmmuPro":null,"scicode":0.478009259259259,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":1000,"tauBanking":0.202061855670103,"mlcrOverall":0.122222222222222,"omniscience":-10.216666666666667,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.19,"gdpPdfAllPass":0.112,"livecodebench":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":2229,"terminalBench21":0.674157303370786,"terminalBench40":0.0101010101010101,"enterpriseOpsGym":0.40226797970754996,"gdpvalNormalized":0.268755,"intelligenceIndex":25.8121062401836,"terminalbenchHard":0.446969696969697,"briefcaseTotalCost":71.7381020277297,"price1mInputTokens":0.95,"contextWindowTokens":256000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.39566666666666667,"price1mOutputTokens":4,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":2.475,"price1mBlended0To3To1":1.7125,"price1mBlended7To2To1":0.7230000000000001,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.9801980198019802,"price1mBlended100To1To1":0.23480392156862745,"automationBenchPartialScore":0.24454304615216713,"intelligenceIndexTimePerTask":591.8122148045513,"omniscienceHallucinationRate":0.8237727523441809,"inferenceParametersActiveBillions":32},"name":"Kimi K2.7 Code","slug":"kimi-k2-7-code","source":"https://artificialanalysis.ai/models/kimi-k2-7-code","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/moonshotai/kimi-k2.7-code/benchmarks"},"created":0,"description":"Kimi K2.7 Code is Moonshot AI's coding specialist for completing software engineering tasks across a codebase. It combines image understanding and tool use with reasoning that remains enabled during generation.","description_sources":[{"url":"https://platform.kimi.ai/docs/guide/kimi-k2-7-code-quickstart","title":"Moonshot model documentation"},{"url":"https://fireworks.ai/models/fireworks/kimi-k2p7-code","title":"Fireworks model documentation"}],"display_name":"Kimi K2.7 Code","context_window":262144,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.95,"output_per_mtok_usd":4,"cached_input_per_mtok_usd":0.19}},{"id":"meituan/longcat-2.0","providers":["atlascloud"],"provider_details":[{"provider":"atlascloud","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/meituan/longcat-2.0/benchmarks"},"created":0,"description":"LongCat 2.0 is Meituan's language model for code understanding, repository changes, and tasks that require multiple steps. Its sparse mixture-of-experts architecture supports text generation and tool use for coding assistants and other agents.","description_sources":[{"url":"https://huggingface.co/meituan-longcat/LongCat-2.0","title":"Meituan model card"},{"url":"https://openrouter.ai/meituan/longcat-2.0","title":"OpenRouter model listing"}],"display_name":"LongCat 2.0","context_window":1048756,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.3,"output_per_mtok_usd":1.2,"cached_input_per_mtok_usd":0.006}},{"id":"xiaomi/mimo-v2.6-pro","providers":["tencent"],"provider_details":[{"provider":"tencent","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":44.8,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":53.2,"artificial_analysis_engineering_index":46.9,"artificial_analysis_intelligence_index":46.3,"artificial_analysis_strategy_and_ops_index":47.8,"artificial_analysis_finance_and_accounting_index":46.1,"artificial_analysis_healthcare_and_medical_index":41.8}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"hle":0.493512511584801,"lcr":0.863333333333333,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.265714285714286,"gdpval":1683.29,"suffix":null,"aa3dElo":980.1231,"ifbench":null,"mmmuPro":null,"scicode":0.608796296296296,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":1020,"tauBanking":null,"mlcrOverall":0.127777777777778,"omniscience":8.383333333333333,"analystAgent":0.375,"frontierCode":0.454153333333333,"cacheHitPrice":0.0036,"gdpPdfAllPass":0.192,"livecodebench":null,"hostModelCount":5,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.348484848484849,"gdpvalNormalized":0.591645,"intelligenceIndex":46.3242065310383,"terminalbenchHard":null,"briefcaseTotalCost":34.596611763440706,"price1mInputTokens":0.435,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.3485,"price1mOutputTokens":0.87,"pricePer1k1mpImages":null,"terminalBenchScience":0.0571428571428571,"price1mBlended0To1To1":0.6525,"price1mBlended0To3To1":0.54375,"price1mBlended7To2To1":0.17652,"cacheHitDiscountPercent":0.9917241379310345,"price1mBlended0To100To1":0.4393069306930693,"price1mBlended100To1To1":0.016323529411764705,"automationBenchPartialScore":0.5862340943419085,"intelligenceIndexTimePerTask":1265.3146509650578,"omniscienceHallucinationRate":0.40624200562803786,"inferenceParametersActiveBillions":42},"name":"MiMo-V2.6-Pro","slug":"mimo-v2-6-pro","source":"https://artificialanalysis.ai/models/mimo-v2-6-pro","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/xiaomi/mimo-v2.6-pro/benchmarks"},"created":0,"description":"MiMo-V2.6-Pro is Xiaomi's most capable reasoning model, built for complex engineering, long-horizon agent tasks and research work, with a one-million-token context window. It reasons by default with an off switch, supports tool calling, and accepts image as well as text input.","description_sources":[{"url":"https://www.tencentcloud.com/document/product/1300/78937","title":"Tencent Cloud TokenHub pricing"},{"url":"https://openrouter.ai/xiaomi/mimo-v2.6-pro","title":"OpenRouter model listing"}],"display_name":"MiMo V2.6 Pro","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.435,"output_per_mtok_usd":0.87,"cached_input_per_mtok_usd":0.0036}},{"id":"xiaomi/mimo-v2.6-flash","providers":["tencent"],"provider_details":[{"provider":"tencent","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":37.2,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":43.6,"artificial_analysis_engineering_index":37.3,"artificial_analysis_intelligence_index":37.9,"artificial_analysis_strategy_and_ops_index":46.1,"artificial_analysis_finance_and_accounting_index":40.3,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"hle":0.350787766450417,"lcr":0.743333333333333,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.12,"gdpval":1607.22,"suffix":null,"ifbench":null,"mmmuPro":0.730635838150289,"scicode":0.512731481481482,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"parameters":309,"tauBanking":null,"mlcrOverall":0.116666666666667,"omniscience":-12.683333333333334,"analystAgent":null,"frontierCode":0.420675333333333,"cacheHitPrice":0.0028,"gdpPdfAllPass":0.09,"livecodebench":null,"hostModelCount":5,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.227272727272727,"gdpvalNormalized":0.55361,"intelligenceIndex":37.8843590141754,"terminalbenchHard":null,"briefcaseTotalCost":13.806623258609111,"price1mInputTokens":0.14,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.27,"price1mOutputTokens":0.28,"pricePer1k1mpImages":null,"terminalBenchScience":0.0428571428571429,"price1mBlended0To1To1":0.21000000000000002,"price1mBlended0To3To1":0.17500000000000002,"price1mBlended7To2To1":0.05796000000000001,"cacheHitDiscountPercent":0.98,"price1mBlended0To100To1":0.1413861386138614,"price1mBlended100To1To1":0.006862745098039216,"automationBenchPartialScore":0.6405168250285657,"intelligenceIndexTimePerTask":1534.7683529421506,"omniscienceHallucinationRate":0.5436073059360731,"inferenceParametersActiveBillions":15},"name":"MiMo-V2.6-Flash","slug":"mimo-v2-6-flash","source":"https://artificialanalysis.ai/models/mimo-v2-6-flash","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/xiaomi/mimo-v2.6-flash/benchmarks"},"created":0,"description":"MiMo-V2.6-Flash is Xiaomi's efficient reasoning model for high-volume, latency-sensitive work, pairing a one-million-token context window with low cost. It reasons by default with an off switch, supports tool calling, and accepts image as well as text input.","description_sources":[{"url":"https://www.tencentcloud.com/document/product/1300/78937","title":"Tencent Cloud TokenHub pricing"},{"url":"https://openrouter.ai/xiaomi/mimo-v2.6-flash","title":"OpenRouter model listing"}],"display_name":"MiMo V2.6 Flash","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.14,"output_per_mtok_usd":0.28,"cached_input_per_mtok_usd":0.0028}},{"id":"xiaomi/mimo-v2.5-pro","providers":["tencent","atlascloud"],"provider_details":[{"provider":"tencent","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]},{"provider":"atlascloud","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":29.6,"artificial_analysis_coding_index":60.2,"artificial_analysis_agentic_index":21.3,"artificial_analysis_economics_index":40.5,"artificial_analysis_engineering_index":28.6,"artificial_analysis_intelligence_index":26,"artificial_analysis_strategy_and_ops_index":20.2,"artificial_analysis_finance_and_accounting_index":27,"artificial_analysis_healthcare_and_medical_index":25.5}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"hle":0.356811862835959,"lcr":0.796666666666667,"gpqa":0.865656565656566,"tau2":0.941520467836257,"aime25":null,"critpt":0.04,"gdpval":1120.77,"ifbench":0.798639455782313,"mmmuPro":null,"scicode":0.505787037037037,"apexAgents":0.0243362831858407,"hleDiamond":null,"itBenchSre":0.382297551789077,"parameters":1023,"tauBanking":0.0989690721649484,"mlcrOverall":0.0555555555555556,"omniscience":3.25,"analystAgent":0.2,"frontierCode":null,"cacheHitPrice":0.0036,"gdpPdfAllPass":0.04,"livecodebench":null,"hostModelCount":6,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.651685393258427,"terminalBench40":0,"gdpvalNormalized":0.31038499999999997,"intelligenceIndex":25.9868888896511,"terminalbenchHard":0.431818181818182,"briefcaseTotalCost":6.538234488024752,"price1mInputTokens":0.435,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.22416666666666665,"price1mOutputTokens":0.87,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.6525,"price1mBlended0To3To1":0.54375,"price1mBlended7To2To1":0.17652,"cacheHitDiscountPercent":0.9917241379310345,"price1mBlended0To100To1":0.4393069306930693,"price1mBlended100To1To1":0.016323529411764705,"automationBenchPartialScore":0.13703616508004313,"intelligenceIndexTimePerTask":678.8099456328255,"omniscienceHallucinationRate":0.24704618689581095,"inferenceParametersActiveBillions":42},"name":"MiMo-V2.5-Pro (Reasoning)","slug":"mimo-v2-5-pro","source":"https://artificialanalysis.ai/models/mimo-v2-5-pro","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/xiaomi/mimo-v2.5-pro/benchmarks"},"created":0,"description":"MiMo V2.5 Pro is Xiaomi's text model for complex software engineering and agent tasks that require sustained instruction following. It is designed to carry work through repeated tool calls and maintain coherence across extended interactions.","description_sources":[{"url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.5-Pro","title":"Xiaomi model card"},{"url":"https://openrouter.ai/xiaomi/mimo-v2.5-pro","title":"OpenRouter model listing"}],"display_name":"MiMo V2.5 Pro","context_window":1024000,"reasoning":null,"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":0.435,"output_per_mtok_usd":0.87,"cached_input_per_mtok_usd":0.004}},{"id":"google/gemini-nano-banana-2.1","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/google/gemini-nano-banana-2.1/benchmarks"},"created":0,"released_at":"2026-10-06","release_source":"https://ai.google.dev/gemini-api/docs/changelog","description":"Nano Banana 2.1 is Google's Flash-tier image generation and conversational editing model, the successor to Nano Banana 2. It is built for high-volume creative work where visual quality, text rendering and keeping a character consistent across several turns of editing matter.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/image-generation","title":"Google image generation documentation"},{"url":"https://ai.google.dev/gemini-api/docs/changelog","title":"Gemini API release notes"},{"url":"https://openrouter.ai/google/gemini-nano-banana-2.1","title":"OpenRouter model listing"}],"display_name":"Nano Banana 2.1","context_window":65536,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":1.5,"output_per_mtok_usd":30,"text_output_per_mtok_usd":7.5}},{"id":"google/gemini-3.1-flash-lite-image","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/google/gemini-3.1-flash-lite-image/benchmarks"},"created":0,"description":"Gemini 3.1 Flash Lite Image, also called Nano Banana 2 Lite, is Google's speed-focused image generation model. It turns text prompts into images for rapid visual exploration and applications that generate many assets.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/image-generation","title":"Google image generation documentation"},{"url":"https://openrouter.ai/google/gemini-3.1-flash-lite-image","title":"OpenRouter model listing"}],"display_name":"Gemini 3.1 Flash Lite Image","context_window":65536,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.25,"output_per_mtok_usd":30,"text_output_per_mtok_usd":1.5}},{"id":"google/gemini-3.1-flash-image","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/google/gemini-3.1-flash-image/benchmarks"},"created":0,"description":"Gemini 3.1 Flash Image, also called Nano Banana 2, is Google's general-purpose image generation model. It combines prompt understanding with text rendering for creating illustrations, diagrams, and other visual assets.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/image-generation","title":"Google image generation documentation"},{"url":"https://openrouter.ai/google/gemini-3.1-flash-image","title":"OpenRouter model listing"}],"display_name":"Gemini 3.1 Flash Image","context_window":65536,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.5,"output_per_mtok_usd":60,"text_output_per_mtok_usd":3}},{"id":"google/gemini-3-pro-image","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/google/gemini-3-pro-image/benchmarks"},"created":0,"description":"Gemini 3 Pro Image, also called Nano Banana Pro, is Google's image generation model for complex visual compositions. It uses reasoning to interpret detailed text prompts and produce images with precise layouts and readable lettering.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/image-generation","title":"Google image generation documentation"},{"url":"https://openrouter.ai/google/gemini-3-pro-image","title":"OpenRouter model listing"}],"display_name":"Gemini 3 Pro Image","context_window":131072,"reasoning":null,"output_modality":"image","output_modalities":["image"],"endpoints":["/v1/images/generations"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":2,"output_per_mtok_usd":120,"text_output_per_mtok_usd":12}},{"id":"google/gemini-embedding-2","providers":["google"],"provider_details":[{"provider":"google","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/google/gemini-embedding-2/benchmarks"},"created":0,"description":"Gemini Embedding 2 is Google's embedding model for representing content as vectors that capture meaning. Its text embeddings support semantic search, document retrieval, classification, and clustering across languages.","description_sources":[{"url":"https://ai.google.dev/gemini-api/docs/embeddings","title":"Google embedding documentation"},{"url":"https://openrouter.ai/google/gemini-embedding-2","title":"OpenRouter model listing"}],"display_name":"Gemini Embedding 2","context_window":8192,"reasoning":null,"output_modality":"embedding","output_modalities":["embedding"],"endpoints":["/v1/embeddings"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.2,"output_per_mtok_usd":0}},{"id":"qwen/qwen3-embedding-8b","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/qwen/qwen3-embedding-8b/benchmarks"},"created":0,"description":"Qwen3 Embedding 8B is Alibaba's text embedding model for search and retrieval applications. It represents queries and documents as vectors so applications can compare meaning across multilingual text.","description_sources":[{"url":"https://fireworks.ai/models/fireworks/qwen3-embedding-8b","title":"Fireworks model documentation"},{"url":"https://openrouter.ai/qwen/qwen3-embedding-8b","title":"OpenRouter model listing"}],"display_name":"Qwen3 Embedding 8B","context_window":40960,"reasoning":null,"output_modality":"embedding","output_modalities":["embedding"],"endpoints":["/v1/embeddings"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.1,"output_per_mtok_usd":0}},{"id":"qwen/qwen3-reranker-8b","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/qwen/qwen3-reranker-8b/benchmarks"},"created":0,"description":"Qwen3 Reranker 8B is Alibaba's model for scoring how well a document answers a query. It reorders candidate documents by relevance to refine search results and select context for retrieval-augmented generation.","description_sources":[{"url":"https://fireworks.ai/models/fireworks/qwen3-reranker-8b","title":"Fireworks model documentation"},{"url":"https://openrouter.ai/qwen/qwen3-reranker-8b","title":"OpenRouter model listing"}],"display_name":"Qwen3 Reranker 8B","context_window":40960,"reasoning":null,"output_modality":"rerank","output_modalities":["rerank"],"endpoints":["/v1/rerank"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.2,"output_per_mtok_usd":0}},{"id":"openai/whisper-large-v3-turbo","providers":["groq"],"provider_details":[{"provider":"groq","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"aa_wer_index":0.046},"name":"Whisper Large v3 Turbo, OpenAI (speech-to-text)","slug":"speech-to-text:087013a3-3e75-4c4f-b31d-170628834d54","source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/openai/whisper-large-v3-turbo/benchmarks"},"created":0,"description":"Whisper Large V3 Turbo is an optimized version of OpenAI's Whisper model for fast multilingual transcription. It uses a pruned and fine-tuned model to turn recorded speech into text with an emphasis on quick turnaround.","description_sources":[{"url":"https://console.groq.com/docs/speech-to-text","title":"Groq speech-to-text documentation"},{"url":"https://openrouter.ai/openai/whisper-large-v3-turbo","title":"OpenRouter model listing"}],"display_name":"Whisper Large V3 Turbo","output_modality":"audio_transcription","input_modalities":["audio","text"],"output_modalities":["text"],"endpoints":["/v1/audio/transcriptions"],"max_file_size_bytes":26214400,"pricing":{"audio_input_per_minute_usd":0.000666666666666667,"audio_minimum_billing_seconds":10}},{"id":"openai/whisper-large-v3","providers":["groq"],"provider_details":[{"provider":"groq","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-21T17:53:53.058Z"},"data":{"aa_wer_index":0.041},"name":"Whisper Large v3, OpenAI (speech-to-text)","slug":"speech-to-text:4810d50c-9471-43f2-a53c-e68f496e24dd","source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","fetchedAt":"2026-10-21T17:53:53.058Z","pageRecords":[]}],"checkedAt":"2026-10-21T17:53:53.058Z","fetchedAt":"2026-10-21T17:53:53.058Z","nextCheckAt":"2026-10-28T17:53:53.058Z","summary":true,"detailPath":"/v1/models/openai/whisper-large-v3/benchmarks"},"created":0,"description":"Whisper Large V3 is OpenAI's multilingual speech recognition model for turning audio recordings into text. It is the full-size alternative to the Turbo variant for transcription workflows that prioritize recognition accuracy.","description_sources":[{"url":"https://console.groq.com/docs/speech-to-text","title":"Groq speech-to-text documentation"},{"url":"https://openrouter.ai/openai/whisper-large-v3","title":"OpenRouter model listing"}],"display_name":"Whisper Large V3","output_modality":"audio_transcription","input_modalities":["audio","text"],"output_modalities":["text"],"endpoints":["/v1/audio/transcriptions"],"max_file_size_bytes":26214400,"pricing":{"audio_input_per_minute_usd":0.00185,"audio_minimum_billing_seconds":10}},{"id":"xai/grok-stt-1.0","providers":["xai"],"provider_details":[{"provider":"xai","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/xai/grok-stt-1.0/benchmarks"},"created":0,"description":"Grok STT is xAI's speech recognition model for transcribing recorded audio. It converts speech into text and accepts language guidance and vocabulary hints for names or specialist terminology.","description_sources":[{"url":"https://docs.x.ai/developers/model-capabilities/audio/speech-to-text","title":"xAI speech-to-text documentation"},{"url":"https://docs.x.ai/developers/models/speech-to-text","title":"xAI speech-to-text model page"}],"display_name":"Grok Speech to Text 1.0","output_modality":"audio_transcription","input_modalities":["audio","text"],"output_modalities":["text"],"endpoints":["/v1/audio/transcriptions"],"max_file_size_bytes":500000000,"pricing":{"audio_input_per_minute_usd":0.00166666666666667}},{"id":"typesafe-ai/jev","providers":["typesafe"],"provider_details":[{"provider":"typesafe","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/typesafe-ai/jev/benchmarks"},"created":0,"released_at":"2026-09-15","release_source":"https://typesafe.ai/blog/introducing-system-one-models-and-jev","description":"Jev is TypeSafe AI's System One decision model: it evaluates a state against typed yes/no, choice, and score questions and returns calibrated probabilities and confidence instead of generated text. It is built for routing, classification, guardrails, and scoring steps that need a fast, structured judgment your code can act on directly.","description_sources":[{"url":"https://docs.typesafe.ai/api","title":"TypeSafe AI API reference"},{"url":"https://typesafe.ai/blog/introducing-system-one-models-and-jev","title":"Introducing System One models and Jev"}],"display_name":"Jev","context_window":32000,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text","image"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.042,"output_per_mtok_usd":0}},{"id":"convaiinnovations/laya","providers":["impossibl"],"provider_details":[{"provider":"impossibl","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/convaiinnovations/laya/benchmarks"},"created":0,"released_at":"2026-09-18","release_source":"https://huggingface.co/convaiinnovations/laya/commits/main","description":"Laya is Convai Innovations' open-weights System One decision model: it answers typed yes/no, choice and score questions about a piece of state in a single non-autoregressive forward pass, across 100+ languages, and never generates text. It is built for routing, moderation, guardrails and triage steps that want a fast calibrated judgment, and it is strongest when the label space is small.","description_sources":[{"url":"https://huggingface.co/convaiinnovations/laya","title":"Laya model card"},{"url":"https://github.com/NandhaKishorM/laya","title":"Laya on GitHub"}],"display_name":"Laya","context_window":8192,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0,"output_per_mtok_usd":0}},{"id":"convaiinnovations/laya-multilingual","providers":["impossibl"],"provider_details":[{"provider":"impossibl","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/convaiinnovations/laya-multilingual/benchmarks"},"created":0,"released_at":"2026-09-19","release_source":"https://huggingface.co/convaiinnovations/laya-multilingual/commits/main","description":"Laya Multilingual is the multilingual checkpoint of Convai Innovations' System One decision model, pinned rather than auto-selected. Use it when your text is not English: Laya's own router picks a checkpoint by script, so Latin-script languages like Spanish and Portuguese are otherwise answered by the English checkpoint.","description_sources":[{"url":"https://huggingface.co/convaiinnovations/Laya-multilingual","title":"Laya Multilingual model card"},{"url":"https://github.com/NandhaKishorM/laya","title":"Laya on GitHub"}],"display_name":"Laya Multilingual","context_window":8192,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0,"output_per_mtok_usd":0}},{"id":"supersoniclabs/julia-1","providers":["impossibl"],"provider_details":[{"provider":"impossibl","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/supersoniclabs/julia-1/benchmarks"},"created":0,"released_at":"2026-09-23","release_source":"https://huggingface.co/SupersonicLabs/Julia-1/commits/main","description":"Julia 1 is Supersonic Labs' first open-weights decision model: a 144M-parameter multilingual encoder that answers typed yes/no, choice and score questions about a piece of state in one forward pass. It is built for classification and routing over options you supply, and is most accurate when every option carries a short descriptive sentence.","description_sources":[{"url":"https://huggingface.co/SupersonicLabs/Julia-1","title":"Julia 1 model card"},{"url":"https://huggingface.co/jhu-clsp/mmBERT-small","title":"mmBERT-small base model card"}],"display_name":"Julia 1","context_window":8192,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.025,"output_per_mtok_usd":0}},{"id":"typesafe-ai/jev-latest","providers":["typesafe"],"provider_details":[{"provider":"typesafe","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/typesafe-ai/jev-latest/benchmarks"},"created":0,"released_at":"2026-09-15","release_source":"https://typesafe.ai/blog/introducing-system-one-models-and-jev","description":"Jev Latest is TypeSafe AI's rolling alias for the newest stable release of Jev, the System One decision model that answers typed yes/no, choice, and score questions with calibrated probabilities and confidence. Use it to stay on the current model automatically; the response reports the version that actually answered.","description_sources":[{"url":"https://docs.typesafe.ai/models","title":"TypeSafe AI models and aliases"},{"url":"https://typesafe.ai/blog/introducing-system-one-models-and-jev","title":"Introducing System One models and Jev"}],"display_name":"Jev Latest","context_window":32000,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text","image"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.042,"output_per_mtok_usd":0}},{"id":"typesafe-ai/jev-1.14.0","providers":["typesafe"],"provider_details":[{"provider":"typesafe","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T09:47:55.482Z","nextCheckAt":"2026-10-28T09:47:55.482Z","summary":true,"detailPath":"/v1/models/typesafe-ai/jev-1.14.0/benchmarks"},"created":0,"released_at":"2026-10-14","release_source":"https://docs.typesafe.ai/models","description":"Jev 1.14.0 is the newest release of TypeSafe AI's System One decision model, which evaluates a state against typed yes/no, choice, and score questions and returns calibrated probabilities and confidence. TypeSafe publishes it as its preview release while jev-latest stays on 1.13.0, so pin it to evaluate against the newest version before it becomes the stable default.","description_sources":[{"url":"https://docs.typesafe.ai/models","title":"TypeSafe AI models and aliases"},{"url":"https://docs.typesafe.ai/api","title":"TypeSafe AI System One API"}],"display_name":"Jev 1.14.0","context_window":32000,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text","image"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.042,"output_per_mtok_usd":0}},{"id":"typesafe-ai/jev-1.13.0","providers":["typesafe"],"provider_details":[{"provider":"typesafe","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"unsupported","variants":[],"checkedAt":"2026-10-27T21:09:52.480Z","nextCheckAt":"2026-10-28T21:09:52.480Z","summary":true,"detailPath":"/v1/models/typesafe-ai/jev-1.13.0/benchmarks"},"created":0,"released_at":"2026-09-15","release_source":"https://typesafe.ai/blog/introducing-system-one-models-and-jev","description":"Jev 1.13.0 is the pinned release of TypeSafe AI's System One decision model, which evaluates a state against typed yes/no, choice, and score questions and returns calibrated probabilities and confidence. Pin it when confidence thresholds are tuned to a specific version and a TypeSafe release should not move them.","description_sources":[{"url":"https://docs.typesafe.ai/models","title":"TypeSafe AI models and aliases"},{"url":"https://docs.typesafe.ai/model-jaggedness/jev-1.13","title":"Jev 1.13 jaggedness"}],"display_name":"Jev 1.13.0","context_window":32000,"reasoning":null,"output_modality":"evaluation","output_modalities":["evaluation"],"endpoints":["/v1/systemone"],"input_modalities":["text","image"],"supports_tools":false,"pricing":{"input_per_mtok_usd":0.042,"output_per_mtok_usd":0}},{"id":"elevenlabs/scribe-v2","providers":["elevenlabs"],"provider_details":[{"provider":"elevenlabs","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":null},"source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","status":"ok","metadata":{"tier":"free"},"fetchedAt":"2026-10-22T18:41:51.835Z"},"data":{"aa_wer_index":0.022},"name":"Scribe v2, ElevenLabs (speech-to-text)","slug":"speech-to-text:69a3c281-fbaa-4102-b7de-7eb99354a561","source":"https://artificialanalysis.ai/api/v2/media/speech-to-text/models/free","fetchedAt":"2026-10-22T18:41:51.835Z","pageRecords":[]}],"checkedAt":"2026-10-22T18:41:51.835Z","fetchedAt":"2026-10-22T18:41:51.835Z","nextCheckAt":"2026-10-29T18:41:51.835Z","summary":true,"detailPath":"/v1/models/elevenlabs/scribe-v2/benchmarks"},"created":0,"description":"Scribe v2 is ElevenLabs' speech recognition model for transcribing recorded audio and video. It returns word-level timestamps, tags audio events such as laughter, and can label which speaker is talking across more than ninety languages.","description_sources":[{"url":"https://elevenlabs.io/docs/api-reference/speech-to-text/convert","title":"ElevenLabs speech-to-text documentation"},{"url":"https://elevenlabs.io/pricing/api","title":"ElevenLabs API pricing"}],"display_name":"Scribe v2","output_modality":"audio_transcription","input_modalities":["audio"],"output_modalities":["text"],"endpoints":["/v1/audio/transcriptions"],"max_file_size_bytes":5000000000,"pricing":{"audio_input_per_minute_usd":0.00366666666666667}},{"id":"elevenlabs/scribe-v2-medical","providers":["elevenlabs"],"provider_details":[{"provider":"elevenlabs","options":[{"region":null,"context_window":null,"supports_tools":false,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:40:53.993Z","nextCheckAt":"2026-10-28T21:40:53.993Z","summary":true,"detailPath":"/v1/models/elevenlabs/scribe-v2-medical/benchmarks"},"created":0,"description":"Scribe v2 Medical is ElevenLabs' speech recognition model tuned for clinical audio. It keeps the Scribe v2 API and languages while cutting errors on medical terminology, and is offered for HIPAA-eligible workloads.","description_sources":[{"url":"https://elevenlabs.io/docs/overview/models","title":"ElevenLabs model overview"},{"url":"https://elevenlabs.io/pricing/api","title":"ElevenLabs API pricing"}],"display_name":"Scribe v2 Medical","output_modality":"audio_transcription","input_modalities":["audio"],"output_modalities":["text"],"endpoints":["/v1/audio/transcriptions"],"max_file_size_bytes":5000000000,"pricing":{"audio_input_per_minute_usd":0.00366666666666667}},{"id":"mistral/mistral-large-4","providers":["mistral"],"provider_details":[{"provider":"mistral","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":37,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":43.6,"artificial_analysis_engineering_index":37.6,"artificial_analysis_intelligence_index":38.4,"artificial_analysis_strategy_and_ops_index":42.2,"artificial_analysis_finance_and_accounting_index":38.3,"artificial_analysis_healthcare_and_medical_index":34.9}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T18:41:51.835Z"},"data":{"hle":0.35032437442076,"lcr":0.813333333333333,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.105714285714286,"gdpval":1420.33,"suffix":null,"aa3dElo":869.7176,"ifbench":null,"mmmuPro":0.764161849710983,"scicode":0.541666666666667,"harveyLab":0.030555555555555555,"apexAgents":null,"hleDiamond":null,"itBenchSre":null,"licenseUrl":null,"parameters":1000,"tauBanking":null,"licenseName":null,"mlcrOverall":0.0777777777777778,"omniscience":-5.3,"analystAgent":null,"frontierCode":0.415277666666667,"cacheHitPrice":0.14,"gdpPdfAllPass":0.186,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":1.36,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.267676767676768,"gdpvalNormalized":0.460165,"intelligenceIndex":38.3770408917494,"terminalbenchHard":null,"briefcaseTotalCost":216.5425582787053,"price1mInputTokens":1.36,"contextWindowTokens":524288,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.25816666666666666,"price1mOutputTokens":4.18,"pricePer1k1mpImages":null,"terminalBenchScience":0.0476190476190476,"modelWeightsSourceUrl":null,"price1mBlended0To1To1":2.77,"price1mBlended0To3To1":2.065,"price1mBlended7To2To1":0.788,"cacheHitDiscountPercent":0.8970588235294118,"price1mBlended0To100To1":1.387920792079208,"price1mBlended100To1To1":0.19156862745098038,"automationBenchPartialScore":0.5990136246808686,"intelligenceIndexTimePerTask":818.1615338343075,"omniscienceHallucinationRate":0.4194563019546169,"inferenceParametersActiveBillions":49},"name":"Mistral Large 4 Preview","slug":"mistral-large-4","source":"https://artificialanalysis.ai/models/mistral-large-4","fetchedAt":"2026-10-22T18:41:51.835Z","pageRecords":[]}],"checkedAt":"2026-10-22T18:41:51.835Z","fetchedAt":"2026-10-22T18:41:51.835Z","nextCheckAt":"2026-10-29T18:41:51.835Z","summary":true,"detailPath":"/v1/models/mistral/mistral-large-4/benchmarks"},"created":0,"released_at":"2026-10-06","release_source":"https://docs.mistral.ai/resources/changelogs","description":"Mistral Large 4 is Mistral's open-weight, general-purpose multimodal model, a granular mixture-of-experts with 52B active and 1.05T total parameters plus a 1.6B vision encoder. It takes text and image input over a long context, can reason before answering or answer directly, and supports function calling and structured outputs.","description_sources":[{"url":"https://docs.mistral.ai/models/mistral-large-4-0","title":"Mistral Large 4 model card"},{"url":"https://docs.mistral.ai/resources/changelogs","title":"Mistral changelog"},{"url":"https://docs.mistral.ai/studio/conversations/reasoning","title":"Mistral reasoning documentation"}],"display_name":"Mistral Large 4","context_window":524288,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1.36,"output_per_mtok_usd":4.18,"cached_input_per_mtok_usd":0.14}},{"id":"fireworks/ember-1","providers":["fireworks"],"provider_details":[{"provider":"fireworks","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-27T21:40:53.993Z","nextCheckAt":"2026-10-28T21:40:53.993Z","summary":true,"detailPath":"/v1/models/fireworks/ember-1/benchmarks"},"created":0,"released_at":"2026-09-23","release_source":"https://fireworks.ai/blog/ember-1","description":"Ember-1 is Fireworks Research's post-trained version of Kimi K3, tuned to reason in shorter traces at comparable quality, which Fireworks reports as roughly 40% fewer output tokens. It keeps Kimi K3's long context, image input and tool use, and can reason before answering or answer directly.","description_sources":[{"url":"https://fireworks.ai/blog/ember-1","title":"Introducing Ember-1"},{"url":"https://fireworks.ai/models/fireworks/ember-1","title":"Fireworks model page"}],"display_name":"Ember-1","context_window":1048576,"reasoning":{"kind":"toggle","levels":[],"off":true,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":3,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.3}},{"id":"stepfun/step-5-preview","providers":["stepfun"],"provider_details":[{"provider":"stepfun","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":["step-5-preview: AA page HTTP 404"],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":46.2,"artificial_analysis_coding_index":null,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":52.8,"artificial_analysis_engineering_index":45,"artificial_analysis_intelligence_index":43.7,"artificial_analysis_strategy_and_ops_index":47.1,"artificial_analysis_finance_and_accounting_index":45.4,"artificial_analysis_healthcare_and_medical_index":41.1}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-27T21:40:53.993Z"},"data":{"hle":0.464782205746061,"lcr":0.883333333333333,"gpqa":null,"tau2":null,"aime25":null,"critpt":0.208571428571429,"gdpval":1594.09,"suffix":null,"aa3dElo":853.5147,"ifbench":null,"mmmuPro":0.763583815028902,"scicode":0.58912037037037,"apexAgents":0.379793510324484,"hleDiamond":null,"itBenchSre":0.555555555555556,"parameters":600,"tauBanking":null,"healthbench":null,"mlcrOverall":0.183333333333333,"omniscience":16.383333333333333,"analystAgent":0.35,"frontierCode":0.384270666666667,"cacheHitPrice":0.1,"code2videoElo":820.57,"gdpPdfAllPass":0.148,"livecodebench":null,"frontierCode20":null,"hostModelCount":1,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":null,"terminalBench40":0.333333333333333,"enterpriseOpsGym":0.4717994628469114,"critptAgenticHard":null,"intelligenceIndex":43.7343049141614,"terminalbenchHard":null,"briefcaseTotalCost":215.92946074175984,"price1mInputTokens":1,"contextWindowTokens":1000000,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.41533333333333333,"price1mOutputTokens":2.7,"pricePer1k1mpImages":null,"terminalBenchScience":0.0238095238095238,"price1mBlended0To1To1":1.85,"price1mBlended0To3To1":1.425,"price1mBlended7To2To1":0.54,"cacheHitDiscountPercent":0.9,"healthbenchProfessional":null,"price1mBlended0To100To1":1.0168316831683168,"price1mBlended100To1To1":0.13431372549019607,"automationBenchPartialScore":0.5095659419096288,"intelligenceIndexTimePerTask":315.4360527507401,"omniscienceHallucinationRate":0.4301596351197263,"inferenceParametersActiveBillions":27},"name":"Step 5 Preview","slug":"step-5","source":"https://artificialanalysis.ai/models/step-5","fetchedAt":"2026-10-27T21:40:53.993Z","pageRecords":[]}],"checkedAt":"2026-10-27T21:40:53.993Z","fetchedAt":"2026-10-14T17:57:48.915Z","nextCheckAt":"2026-10-28T21:40:53.993Z","summary":true,"detailPath":"/v1/models/stepfun/step-5-preview/benchmarks"},"created":0,"released_at":"2026-09-20","release_source":"https://platform.stepfun.ai/docs/en/guides/models/step-5-preview","description":"Step 5 Preview is StepFun's flagship model for agentic work, aimed at software engineering and professional knowledge work with particular strength in finance. It takes text and image input over a context of about one million tokens, reasons at a selectable effort level before answering, and supports tool calling and structured output.","description_sources":[{"url":"https://platform.stepfun.ai/docs/en/guides/models/step-5-preview","title":"Step 5 Preview model page"},{"url":"https://platform.stepfun.ai/docs/en/guides/developer/reasoning","title":"StepFun reasoning model guide"}],"display_name":"Step 5 Preview","context_window":1024000,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":1,"output_per_mtok_usd":2.7,"cached_input_per_mtok_usd":0.05}},{"id":"stepfun/step-3.7-flash","providers":["stepfun"],"provider_details":[{"provider":"stepfun","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"available","variants":[{"api":{"data":{"evaluations":{"artificial_analysis_legal_index":null,"artificial_analysis_coding_index":39.6,"artificial_analysis_agentic_index":null,"artificial_analysis_economics_index":null,"artificial_analysis_engineering_index":null,"artificial_analysis_intelligence_index":19.5,"artificial_analysis_strategy_and_ops_index":null,"artificial_analysis_finance_and_accounting_index":null,"artificial_analysis_healthcare_and_medical_index":null}},"source":"https://artificialanalysis.ai/api/v2/language/models/free","status":"ok","metadata":{"tier":"free","intelligence_index_version":4.3},"fetchedAt":"2026-10-22T18:41:51.835Z"},"data":{"hle":0.214087117701576,"lcr":0.736666666666667,"gpqa":0.809090909090909,"tau2":0.985380116959064,"aime25":null,"critpt":0.0228571428571429,"gdpval":856.36,"suffix":null,"ifbench":0.672789115646258,"mmmuPro":0.753179190751445,"scicode":0.438657407407407,"apexAgents":0.148230088495575,"hleDiamond":null,"itBenchSre":0.302730696798493,"parameters":198,"tauBanking":0.119587628865979,"omniscience":-37.28333333333333,"analystAgent":null,"frontierCode":null,"cacheHitPrice":0.04,"livecodebench":null,"hostModelCount":2,"cacheWritePrice":null,"reasoningTokens":2000,"terminalBench21":0.393258426966292,"terminalBench40":null,"enterpriseOpsGym":0.29065950462548495,"gdpvalNormalized":0.17818,"intelligenceIndex":19.4812572685302,"terminalbenchHard":0.356060606060606,"price1mInputTokens":0.2,"contextWindowTokens":262144,"knowledgeCutoffDate":null,"omniscienceAccuracy":0.25783333333333336,"price1mOutputTokens":1.15,"pricePer1k1mpImages":null,"terminalBenchScience":null,"price1mBlended0To1To1":0.6749999999999999,"price1mBlended0To3To1":0.4375,"price1mBlended7To2To1":0.183,"cacheHitDiscountPercent":0.8,"price1mBlended0To100To1":0.2094059405940594,"price1mBlended100To1To1":0.05245098039215686,"omniscienceHallucinationRate":0.8497642039074781,"inferenceParametersActiveBillions":11},"name":"Step 3.7 Flash","slug":"step-3-7-flash","source":"https://artificialanalysis.ai/models/step-3-7-flash","fetchedAt":"2026-10-22T18:41:51.835Z","pageRecords":[]}],"checkedAt":"2026-10-22T18:41:51.835Z","fetchedAt":"2026-10-22T18:41:51.835Z","nextCheckAt":"2026-10-29T18:41:51.835Z","summary":true,"detailPath":"/v1/models/stepfun/step-3.7-flash/benchmarks"},"created":0,"released_at":"2026-05-29","release_source":"https://huggingface.co/stepfun-ai/Step-3.7-Flash","description":"Step 3.7 Flash is StepFun's open-weight, high-speed multimodal reasoning model, a sparse mixture-of-experts with 198B total and 11B active parameters. It takes text and image input over a 256K context, reasons at a selectable effort level before answering, and is tuned for agent, coding and tool-calling workloads.","description_sources":[{"url":"https://platform.stepfun.ai/docs/en/guides/models/step-3.7-flash","title":"Step 3.7 Flash model page"},{"url":"https://huggingface.co/stepfun-ai/Step-3.7-Flash","title":"Step 3.7 Flash on Hugging Face"}],"display_name":"Step 3.7 Flash","context_window":262144,"reasoning":{"kind":"effort","levels":["low","medium","high"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text","image"],"pricing":{"input_per_mtok_usd":0.2,"output_per_mtok_usd":1.15,"cached_input_per_mtok_usd":0.04}},{"id":"impossibl/shadowmancer-preview","providers":["impossibl"],"provider_details":[{"provider":"impossibl","options":[{"region":null,"context_window":null,"supports_tools":null,"route_count":1}]}],"object":"model","benchmarks":{"errors":[],"source":"Artificial Analysis","status":"not_found","variants":[],"checkedAt":"2026-10-28T05:25:53.374Z","nextCheckAt":"2026-10-29T05:25:53.374Z","summary":true,"detailPath":"/v1/models/impossibl/shadowmancer-preview/benchmarks"},"created":0,"display_name":"Shadowmancer Preview","context_window":1000000,"reasoning":{"kind":"effort","levels":["high","xhigh"],"off":false,"default":null},"output_modality":"text","output_modalities":["text"],"endpoints":["/v1/chat/completions"],"input_modalities":["text"],"pricing":{"input_per_mtok_usd":3,"output_per_mtok_usd":15,"cached_input_per_mtok_usd":0.5},"invite_only":true}]}