# Impossibl AI API The Impossibl AI API is one API for all AI providers, OpenAI-compatible across OpenAI, Anthropic, Google, xAI, and more (full model list below), billed from a prepaid credit balance. Human sign-in uses WorkOS AuthKit; programmatic access uses gateway API keys minted after account creation. ## Quickstart (agent) 1. Read https://api.impossibl.com/auth.md. 2. For the frictionless pre-claim flow, POST /v1/accounts -> returns an API key (imp-rt-...) and a $0.05 signup bonus. 3. POST /v1/chat/completions, /v1/responses, or /v1/messages with the API key in the Authorization bearer header. 4. Tell the user to claim the account to get the extra credits: an agent-created account has $0.05; claiming adds an extra $1. POST /agent/identity/claim with the user's email, send them the verification_uri_complete link (code pre-filled — they just sign in + confirm), and poll POST /oauth2/token until claimed. ## Discovery - https://api.impossibl.com/.well-known/oauth-protected-resource - https://api.impossibl.com/auth.md - https://api.impossibl.com/.well-known/auth.md ## Endpoints - GET /v1/models list models + pricing (no auth) - POST /v1/accounts create a funded account + first key (no auth) - GET /v1/account balance + identity - POST /v1/keys mint a key - GET /v1/keys list keys - PATCH /v1/keys/{id} rename a key - DELETE /v1/keys/{id} revoke a key - GET /v1/usage recent usage ledger - GET /v1/usage/stream live usage events over sse (snapshot + request) - GET /v1/requests recent request log rows (limit, since, before_id, status, model) - GET /v1/requests/{id} one request + captured payload - GET /v1/billing balance + minimum top-up + fee rate - POST /v1/billing/invoice buy net credits + 5% platform fee before tax - GET /v1/provider-keys list BYOK provider credentials - POST /v1/provider-keys add your own provider key (BYOK, zero-credit routing) - DELETE /v1/provider-keys/{id} revoke a BYOK credential - POST /agent/identity/claim start a claim ceremony (bearer key) -> user_code + link - POST /oauth2/token poll the claim ceremony (claim grant) - POST /v1/chat/completions OpenAI Chat Completions (stream optional) - POST /v1/responses OpenAI Responses API (stream optional) - POST /v1/messages Anthropic Messages API (stream optional) - POST /v1/images/generations OpenAI Images (image-output models only) - POST /v1/embeddings OpenAI Embeddings (embedding models only) - POST /v1/rerank rank documents by query relevance (reranking models only) - POST /v1/systemone TypeSafe System One evaluation: state + typed questions -> typed answers (evaluation models only) - POST /v1/decisions OpenAI Decisions: input + typed questions -> typed answers (decisions models only) - POST /v1/audio/transcriptions OpenAI-compatible file transcription (stream optional) - POST /v1/audio/speech OpenAI-compatible text to speech (returns audio bytes) ## Embeddings ```bash curl https://api.impossibl.com/v1/embeddings \ -H "Authorization: Bearer $AI_API_KEY" \ -H 'content-type: application/json' \ -d '{"model":"openai/text-embedding-3-small","input":"the quick brown fox"}' ``` ## Audio transcription ```bash curl https://api.impossibl.com/v1/audio/transcriptions \ -H "Authorization: Bearer $AI_API_KEY" \ -F file=@/path/to/audio.wav \ -F model=openai/gpt-transcribe ``` ## Text to speech Voices are provider names, listed per model by GET /v1/models. ```bash curl https://api.impossibl.com/v1/audio/speech \ -H "Authorization: Bearer $AI_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"xai/grok-voice-tts-1.0","input":"hello there","voice":"eve"}' \ --output speech.mp3 ``` ## Machine docs - /openapi.yaml OpenAPI 3.1 spec - /llms.txt API quickstart + endpoints + models - /auth.md executable agent onboarding manifest ## Human docs - https://impossibl.com/docs/quickstart create a key in the console and make a first request - https://impossibl.com/docs/integrations SDK and coding-tool setup - https://impossibl.com/docs/api-compatibility supported formats and stateless conversation limits - https://impossibl.com/docs/models model selection and live discovery - https://impossibl.com/docs/billing credits, fees, auto top-up, and balance checks - https://impossibl.com/docs/errors diagnose failures and choose when to retry ## Models (use the full provider/model id) - openai/gpt-6-astra — GPT-6 Astra, 1,050,000 ctx, $10/$50 per 1M in/out tokens - openai/gpt-6.1-sol — GPT-6.1 Sol, 1,050,000 ctx, $2/$10 per 1M in/out tokens - openai/gpt-6-sol — GPT-6 Sol, 1,050,000 ctx, $2/$10 per 1M in/out tokens - openai/gpt-6-luna — GPT-6 Luna, 1,050,000 ctx, $0.1/$0.5 per 1M in/out tokens - openai/gpt-6-luna-decisions — GPT-6 Luna Decisions, evaluation (typed questions -> typed answers), $0.1 per 1M input tokens - openai/gpt-5.6-sol — GPT-5.6 Sol, 1,050,000 ctx, $4/$20 per 1M in/out tokens - openai/gpt-5.6-luna — GPT-5.6 Luna, 1,050,000 ctx, $0.2/$1.2 per 1M in/out tokens - openai/gpt-5.6-terra — GPT-5.6 Terra, 1,050,000 ctx, $2/$12 per 1M in/out tokens - openai/gpt-image-2.5-sunburst — GPT Image 2.5 Sunburst, 32,000 ctx, $5/$30 per 1M in/out tokens - openai/gpt-image-2.5-flare — GPT Image 2.5 Flare, 32,000 ctx, $5/$30 per 1M in/out tokens - openai/gpt-image-2 — GPT Image 2, 32,000 ctx, $5/$30 per 1M in/out tokens - openai/gpt-transcribe - GPT Transcribe, $0.0045 per audio minute - openai/text-embedding-3-small — Text Embedding 3 Small, 8,192 ctx, $0.02/$0 per 1M in/out tokens - openai/text-embedding-3-large — Text Embedding 3 Large, 8,192 ctx, $0.13/$0 per 1M in/out tokens - openai/gpt-5.5 — GPT-5.5, 1,050,000 ctx, $5/$30 per 1M in/out tokens - openai/gpt-5.5-pro — GPT-5.5 Pro, 1,050,000 ctx, $30/$180 per 1M in/out tokens - openai/gpt-5.4 — GPT-5.4, 1,050,000 ctx, $2.5/$15 per 1M in/out tokens - openai/gpt-5.4-mini — GPT-5.4 mini, 400,000 ctx, $0.75/$4.5 per 1M in/out tokens - openai/gpt-5.4-nano — GPT-5.4 nano, 400,000 ctx, $0.2/$1.25 per 1M in/out tokens - openai/gpt-5.4-pro — GPT-5.4 Pro, 1,050,000 ctx, $30/$180 per 1M in/out tokens - openai/gpt-5.3-codex — GPT-5.3 Codex, 400,000 ctx, $1.75/$14 per 1M in/out tokens - openai/gpt-5.1 — GPT-5.1, 400,000 ctx, $1.25/$10 per 1M in/out tokens - openai/gpt-5.2 — GPT-5.2, 400,000 ctx, $1.75/$14 per 1M in/out tokens - openai/gpt-5.2-codex — GPT-5.2 Codex, 400,000 ctx, $1.75/$14 per 1M in/out tokens - openai/gpt-5.1-codex — GPT-5.1 Codex, 400,000 ctx, $1.25/$10 per 1M in/out tokens - openai/gpt-5.1-codex-mini — GPT-5.1 Codex mini, 400,000 ctx, $0.25/$2 per 1M in/out tokens - openai/chat-latest — GPT Chat Latest, 400,000 ctx, $5/$30 per 1M in/out tokens - openai/gpt-5 — GPT-5, 400,000 ctx, $1.25/$10 per 1M in/out tokens - openai/gpt-5-pro — GPT-5 Pro, 400,000 ctx, $15/$120 per 1M in/out tokens - openai/gpt-5-codex — GPT-5 Codex, 400,000 ctx, $1.25/$10 per 1M in/out tokens - openai/gpt-5-mini — GPT-5 mini, 400,000 ctx, $0.25/$2 per 1M in/out tokens - openai/gpt-5-nano — GPT-5 nano, 400,000 ctx, $0.05/$0.4 per 1M in/out tokens - openai/gpt-4.1 — GPT-4.1, 1,000,000 ctx, $2/$8 per 1M in/out tokens - openai/gpt-4.1-mini — GPT-4.1 mini, 1,000,000 ctx, $0.4/$1.6 per 1M in/out tokens - openai/gpt-4.1-nano — GPT-4.1 nano, 1,000,000 ctx, $0.1/$0.4 per 1M in/out tokens - openai/gpt-4o — GPT-4o, 128,000 ctx, $2.5/$10 per 1M in/out tokens - openai/gpt-4o-mini — GPT-4o mini, 128,000 ctx, $0.15/$0.6 per 1M in/out tokens - openai/gpt-4-turbo — GPT-4 Turbo, 128,000 ctx, $10/$30 per 1M in/out tokens - openai/gpt-3.5-turbo — GPT-3.5 Turbo, 16,000 ctx, $0.5/$1.5 per 1M in/out tokens - openai/o3 — OpenAI o3, 200,000 ctx, $2/$8 per 1M in/out tokens - openai/o3-mini — OpenAI o3-mini, 200,000 ctx, $1.1/$4.4 per 1M in/out tokens - openai/o4-mini — OpenAI o4-mini, 200,000 ctx, $1.1/$4.4 per 1M in/out tokens - openai/o1 — OpenAI o1, 200,000 ctx, $15/$60 per 1M in/out tokens - openai/gpt-oss-120b — GPT OSS 120B, 128,000 ctx, $0.35/$0.75 per 1M in/out tokens - openai/gpt-oss-20b — GPT OSS 20B, 128,000 ctx, $0.075/$0.3 per 1M in/out tokens - anthropic/claude-opus-5 — Claude Opus 5, 1,000,000 ctx, $5/$25 per 1M in/out tokens - anthropic/claude-opus-5-5 — Claude Opus 5.5, 1,000,000 ctx, $4/$20 per 1M in/out tokens - anthropic/claude-fable-5-1 — Claude Fable 5.1, 1,000,000 ctx, $10/$50 per 1M in/out tokens - anthropic/claude-fable-5 — Claude Fable 5, 1,000,000 ctx, $10/$50 per 1M in/out tokens - anthropic/claude-opus-4-8 — Claude Opus 4.8, 1,000,000 ctx, $5/$25 per 1M in/out tokens - anthropic/claude-opus-4-7 — Claude Opus 4.7, 1,000,000 ctx, $5/$25 per 1M in/out tokens - anthropic/claude-opus-4-6 — Claude Opus 4.6, 1,000,000 ctx, $5/$25 per 1M in/out tokens - anthropic/claude-sonnet-5-5 — Claude Sonnet 5.5, 1,000,000 ctx, $2/$10 per 1M in/out tokens - anthropic/claude-sonnet-5 — Claude Sonnet 5, 1,000,000 ctx, $2/$10 per 1M in/out tokens - anthropic/claude-sonnet-4-6 — Claude Sonnet 4.6, 1,000,000 ctx, $3/$15 per 1M in/out tokens - anthropic/claude-haiku-5-5 — Claude Haiku 5.5, 1,000,000 ctx, $0.1/$0.5 per 1M in/out tokens - anthropic/claude-haiku-4-5 — Claude Haiku 4.5, 200,000 ctx, $1/$5 per 1M in/out tokens - anthropic/claude-opus-4-5 — Claude Opus 4.5, 200,000 ctx, $5/$25 per 1M in/out tokens - anthropic/claude-sonnet-4-5 — Claude Sonnet 4.5, 200,000 ctx, $3/$15 per 1M in/out tokens - google/gemini-4-argon — Gemini 4 Argon, 1,048,576 ctx, $2/$10 per 1M in/out tokens - google/gemini-3.8-flash — Gemini 3.8 Flash, 1,048,576 ctx, $0.75/$3.75 per 1M in/out tokens - google/gemini-3.7-flash — Gemini 3.7 Flash, 1,048,576 ctx, $0.75/$3.75 per 1M in/out tokens - google/gemini-robotics-er-2-preview — Gemini Robotics-ER 2 Preview, 131,072 ctx, $1/$5 per 1M in/out tokens - google/gemini-3.6-flash — Gemini 3.6 Flash, 1,048,576 ctx, $0.75/$3.75 per 1M in/out tokens - google/gemini-3.5-flash — Gemini 3.5 Flash, 1,000,000 ctx, $1.5/$9 per 1M in/out tokens - google/gemini-3.5-flash-lite — Gemini 3.5 Flash-Lite, 1,048,576 ctx, $0.3/$2.5 per 1M in/out tokens - google/gemini-3.1-pro-preview — Gemini 3.1 Pro (preview), 1,000,000 ctx, $2/$12 per 1M in/out tokens - google/gemini-3.1-flash-lite — Gemini 3.1 Flash-Lite, 1,000,000 ctx, $0.25/$1.5 per 1M in/out tokens - google/gemini-2.5-pro — Gemini 2.5 Pro, 1,000,000 ctx, $1.25/$10 per 1M in/out tokens - google/gemini-2.5-flash — Gemini 2.5 Flash, 1,000,000 ctx, $0.3/$2.5 per 1M in/out tokens - google/gemini-2.5-flash-lite — Gemini 2.5 Flash-Lite, 1,000,000 ctx, $0.1/$0.4 per 1M in/out tokens - google/gemma-4-31b — Gemma 4 31B, 128,000 ctx, $0.99/$1.49 per 1M in/out tokens - google/gemini-3.8-flash-tts - Gemini 3.8 Flash TTS, $1/$18 per 1M input/output tokens (speech) - google/gemini-3.1-flash-tts-preview - Gemini 3.1 Flash TTS (preview), $1/$20 per 1M input/output tokens (speech) - xai/grok-4.7 — Grok 4.7, 500,000 ctx, $2/$6 per 1M in/out tokens - xai/grok-4.6 — Grok 4.6, 500,000 ctx, $2/$6 per 1M in/out tokens - xai/grok-4.5 — Grok 4.5, 500,000 ctx, $2/$6 per 1M in/out tokens - xai/grok-4.3 — Grok 4.3, 1,000,000 ctx, $1.25/$2.5 per 1M in/out tokens - xai/grok-4.20-0309-reasoning — Grok 4.20 (reasoning), 1,000,000 ctx, $1.25/$2.5 per 1M in/out tokens - xai/grok-4.20-0309-non-reasoning — Grok 4.20 (non-reasoning), 1,000,000 ctx, $1.25/$2.5 per 1M in/out tokens - xai/grok-4.20-multi-agent-0309 — Grok 4.20 (multi-agent), 1,000,000 ctx, $1.25/$2.5 per 1M in/out tokens - xai/grok-build-0.1 — Grok Build 0.1, 256,000 ctx, $1/$2 per 1M in/out tokens - xai/grok-voice-tts-1.0 - Grok Voice TTS 1.0, $15 per 1M characters - meta/llama-3.3-70b — Llama 3.3 70B Instruct, 128,000 ctx, $0.71/$0.71 per 1M in/out tokens - meta/muse-glimmer-30b — Muse Glimmer 30B, 131,072 ctx, $0.35/$1.5 per 1M in/out tokens - mistral/mistral-large-3 — Mistral Large 3, 128,000 ctx, $0.5/$1.5 per 1M in/out tokens - microsoft/phi-4 — Phi-4, 16,000 ctx, $0.125/$0.5 per 1M in/out tokens - deepseek/deepseek-v4-flash-0731 — DeepSeek V4 Flash 0731, 1,048,576 ctx, $0.14/$0.28 per 1M in/out tokens - deepseek/deepseek-v4-pro-0813 — DeepSeek V4 Pro 0813, 1,048,576 ctx, $1.74/$3.48 per 1M in/out tokens - deepseek/deepseek-v4.1-flash — DeepSeek V4.1 Flash, 1,048,576 ctx, $0.3/$1.2 per 1M in/out tokens - nvidia/nemotron-3.5-lightning — Nemotron 3.5 Lightning, 262,144 ctx, $0.05/$0.2 per 1M in/out tokens - deepseek/deepseek-v4-flash — DeepSeek V4 Flash, 1,000,000 ctx, $0.19/$0.51 per 1M in/out tokens - deepseek/deepseek-v4-pro — DeepSeek V4 Pro, 1,000,000 ctx, $1.74/$3.48 per 1M in/out tokens - deepseek/deepseek-v3-2 — DeepSeek V3.2, 163,840 ctx, $0.58/$1.68 per 1M in/out tokens - deepseek/deepseek-v3-2-speciale — DeepSeek V3.2 Speciale, 163,840 ctx, $0.58/$1.68 per 1M in/out tokens - zai/glm-5.3 — GLM-5.3, 1,000,000 ctx, $1.4/$4.4 per 1M in/out tokens - zai/glm-5.3-fast — GLM-5.3 Fast, 1,000,000 ctx, $2.1/$6.6 per 1M in/out tokens - zai/glm-5.3-flash — GLM-5.3-Flash, 1,000,000 ctx, $0.15/$0.5 per 1M in/out tokens - zai/glm-5.3-flashx — GLM-5.3-FlashX, 1,000,000 ctx, $0.37/$1.25 per 1M in/out tokens - zai/glm-5.2 — GLM-5.2, 1,000,000 ctx, $1.4/$4.4 per 1M in/out tokens - zai/glm-5.1 — GLM-5.1, 200,000 ctx, $1.4/$4.4 per 1M in/out tokens - zai/glm-5 — GLM-5, 200,000 ctx, $1/$3.2 per 1M in/out tokens - zai/glm-5-turbo — GLM-5 Turbo, 200,000 ctx, $1.2/$4 per 1M in/out tokens - zai/glm-4.7 — GLM-4.7, 200,000 ctx, $0.6/$2.2 per 1M in/out tokens - zai/glm-4.6 — GLM-4.6, 200,000 ctx, $0.6/$2.2 per 1M in/out tokens - zai/glm-4.5 — GLM-4.5, 128,000 ctx, $0.6/$2.2 per 1M in/out tokens - zai/glm-4.5-air — GLM-4.5 Air, 128,000 ctx, $0.2/$1.1 per 1M in/out tokens - moonshotai/kimi-k3 — Kimi K3, 1,048,576 ctx, $3/$15 per 1M in/out tokens - qwen/qwen3.7-max — Qwen3.7 Max, 1,048,576 ctx, $2.5/$7.5 per 1M in/out tokens - qwen/qwen3.7-plus — Qwen3.7 Plus, 1,048,576 ctx, $0.4/$1.6 per 1M in/out tokens - qwen/qwen3.6-flash — Qwen3.6 Flash, 1,048,576 ctx, $0.25/$1.5 per 1M in/out tokens - qwen/qwen3.6-27b — Qwen3.6 27B, 229,376 ctx, $0.6/$3.6 per 1M in/out tokens - qwen/qwen3.8-max — Qwen3.8 Max, 1,048,576 ctx, $2/$6 per 1M in/out tokens - qwen/qwen3.8-27b — Qwen3.8 27B, 1,048,576 ctx, $0.5/$3 per 1M in/out tokens - qwen/qwen3.8-2.4t-a95b — Qwen3.8 2.4T A95B, 262,144 ctx, $1.2/$1.2 per 1M in/out tokens - xiaomi/mimo-v2.5 — MiMo V2.5, 1,024,000 ctx, $0.14/$0.28 per 1M in/out tokens - minimaxai/minimax-m3 — MiniMax M3, 524,300 ctx, $0.3/$1.2 per 1M in/out tokens - tencent/hy3 — Hunyuan 3, 262,144 ctx, $0.132/$0.528 per 1M in/out tokens - tencent/hy4-preview — Hunyuan 4 Preview, 1,048,576 ctx, $0.834/$2.501 per 1M in/out tokens - thinkingmachines/inkling — Inkling, 65,536 ctx, $1.87/$4.68 per 1M in/out tokens - thinkingmachines/inkling-256k — Inkling 256K, 262,144 ctx, $3.74/$9.36 per 1M in/out tokens - meta/muse-wonder — Muse Wonder, 1,048,576 ctx, $4/$20 per 1M in/out tokens - meta/muse-spark-1.3 — Muse Spark 1.3, 1,048,576 ctx, $1.25/$4.25 per 1M in/out tokens - meta/muse-spark-1.3-contributor — Muse Spark 1.3 (Contributor), 1,048,576 ctx, $0.1/$0.2 per 1M in/out tokens - qwen/qwen3.8-flash — Qwen3.8 Flash, 1,000,000 ctx, $0.15/$0.47 per 1M in/out tokens - qwen/qwen3.7-flash — Qwen3.7 Flash, 1,000,000 ctx, $0.03/$0.13 per 1M in/out tokens - thinkingmachines/inkling-small — Inkling Small, 65,536 ctx, $0.58/$1.44 per 1M in/out tokens - thinkingmachines/inkling-small-256k — Inkling Small 256K, 262,144 ctx, $1.16/$2.89 per 1M in/out tokens - nvidia/nemotron-3-ultra-550b-a55b — Nemotron 3 Ultra 550B A55B, 262,144 ctx, $0.6/$2.4 per 1M in/out tokens - moonshotai/kimi-k2.7-code — Kimi K2.7 Code, 262,144 ctx, $0.95/$4 per 1M in/out tokens - meituan/longcat-2.0 — LongCat 2.0, 1,048,756 ctx, $0.3/$1.2 per 1M in/out tokens - xiaomi/mimo-v2.6-pro — MiMo V2.6 Pro, 1,048,576 ctx, $0.435/$0.87 per 1M in/out tokens - xiaomi/mimo-v2.6-flash — MiMo V2.6 Flash, 1,048,576 ctx, $0.14/$0.28 per 1M in/out tokens - xiaomi/mimo-v2.5-pro — MiMo V2.5 Pro, 1,024,000 ctx, $0.435/$0.87 per 1M in/out tokens - google/gemini-nano-banana-2.1 — Nano Banana 2.1, 65,536 ctx, $1.5/$30 per 1M in/out tokens - google/gemini-3.1-flash-lite-image — Gemini 3.1 Flash Lite Image, 65,536 ctx, $0.25/$30 per 1M in/out tokens - google/gemini-3.1-flash-image — Gemini 3.1 Flash Image, 65,536 ctx, $0.5/$60 per 1M in/out tokens - google/gemini-3-pro-image — Gemini 3 Pro Image, 131,072 ctx, $2/$120 per 1M in/out tokens - google/gemini-embedding-2 — Gemini Embedding 2, 8,192 ctx, $0.2/$0 per 1M in/out tokens - qwen/qwen3-embedding-8b — Qwen3 Embedding 8B, 40,960 ctx, $0.1/$0 per 1M in/out tokens - qwen/qwen3-reranker-8b — Qwen3 Reranker 8B, reranking, $0.2 per 1M input tokens - openai/whisper-large-v3-turbo - Whisper Large V3 Turbo, $0.000666666666666667 per audio minute - openai/whisper-large-v3 - Whisper Large V3, $0.00185 per audio minute - xai/grok-stt-1.0 - Grok Speech to Text 1.0, $0.00166666666666667 per audio minute - typesafe-ai/jev — Jev, evaluation (typed questions -> typed answers), $0.042 per 1M input tokens - convaiinnovations/laya — Laya, evaluation (typed questions -> typed answers), $0 per 1M input tokens - convaiinnovations/laya-multilingual — Laya Multilingual, evaluation (typed questions -> typed answers), $0 per 1M input tokens - supersoniclabs/julia-1 — Julia 1, evaluation (typed questions -> typed answers), $0.025 per 1M input tokens - typesafe-ai/jev-latest — Jev Latest, evaluation (typed questions -> typed answers), $0.042 per 1M input tokens - typesafe-ai/jev-1.14.0 — Jev 1.14.0, evaluation (typed questions -> typed answers), $0.042 per 1M input tokens - typesafe-ai/jev-1.13.0 — Jev 1.13.0, evaluation (typed questions -> typed answers), $0.042 per 1M input tokens - elevenlabs/scribe-v2 - Scribe v2, $0.00366666666666667 per audio minute - elevenlabs/scribe-v2-medical - Scribe v2 Medical, $0.00366666666666667 per audio minute - mistral/mistral-large-4 — Mistral Large 4, 524,288 ctx, $0.68/$2.09 per 1M in/out tokens - fireworks/ember-1 — Ember-1, 1,048,576 ctx, $3/$15 per 1M in/out tokens - stepfun/step-5-preview — Step 5 Preview, 1,024,000 ctx, $1/$2.7 per 1M in/out tokens - stepfun/step-3.7-flash — Step 3.7 Flash, 262,144 ctx, $0.2/$1.15 per 1M in/out tokens ## Public model directory - [Browse models](https://impossibl.com/models): Model details, published API prices, capabilities, and comparisons. - [Model directory as Markdown](https://impossibl.com/models.md): Model references and links to each model's Markdown page. - [Compare models as Markdown](https://impossibl.com/compare.md): Starting comparisons from the serving catalog. - [Model catalog as JSON](https://impossibl.com/models.json): Public model facts with canonical page URLs and a markdown_url for each model. - [Model directory sitemap](https://impossibl.com/models/sitemap.xml): Model, creator, and comparison pages. Per-model Markdown is available at `https://impossibl.com//.md`. Follow each model's `markdown_url` in the JSON catalog for its prices, limits, API identifiers, FAQs, and sources. Creator pages are at `https://impossibl.com//`, with Markdown at `https://impossibl.com/.md`. Append `.md` to comparison page URLs for their Markdown representations. These pages also return Markdown at their HTML URLs when requested with `Accept: text/markdown`. The homepage is available as Markdown at [index.md](https://impossibl.com/index.md), or by requesting the homepage with `Accept: text/markdown`. ## Company and updates - [Changelog as Markdown](https://impossibl.com/changelog.md) - [Company as Markdown](https://impossibl.com/about.md) - [Brand assets as Markdown](https://impossibl.com/brand.md) These representations are generated from the same content as their human pages and are also available at the HTML URLs with `Accept: text/markdown`.