# AI Models Router

> One API. Every AI Model. One OpenAI-compatible API for 32 open-weight AI models, paid from prepaid credits at the per-token price shown on each model. Failed requests are not charged.

- Base URL: https://api.aimodelsrouter.ai/v1 (OpenAI Chat Completions format; any OpenAI SDK works)
- Auth: `Authorization: Bearer sk-amr-v1-…`, keys from https://aimodelsrouter.ai/dashboard/keys
- Endpoints: `GET /models` (no auth), `GET /models/{id}`, `POST /chat/completions` (streaming, tool calling, JSON schema, image input on vision models), `POST /responses` (OpenAI Responses API, for Codex; stateless), `POST /messages` and `POST /messages/count_tokens` (Anthropic Messages API, for Claude Code; `x-api-key` or Bearer), `GET /key`, `GET /generation?id=gen-…`, `GET /generations` (your recent requests), `GET /usage` (cost and tokens per day or per model)
- Anthropic clients (Claude Code, Anthropic SDKs) use https://api.aimodelsrouter.ai as base URL, without /v1, and one of our model ids
- Routing controls on every inference endpoint: `max_price` (USD per 1M tokens; refused before anything is reserved if the model costs more) and `zdr: true` (zero-data-retention routes only)
- Not available: embeddings, image or audio generation, stored responses (`previous_response_id`)
- Catalog: 32 open-weight models; use the `id` from `GET /models` as `model`
- Billing: prepaid USD credits. A request holds its maximum possible cost, then is charged its actual cost (`usage.cost` in every response)
- No charge on failure: a request that fails before any output (every route down, timeout, capacity, invalid request, a limit) costs nothing. A stream that breaks after output started is charged only for the prompt and the output delivered
- Limits: 600 requests per minute per key; optional per-key spending, token and concurrency limits
- Machine-readable: OpenAPI 3.1 at https://aimodelsrouter.ai/openapi.json; API catalog (RFC 9727) at https://aimodelsrouter.ai/.well-known/api-catalog

## Docs

- [Quickstart](https://aimodelsrouter.ai/docs): Make your first request in under a minute.
- [Authentication](https://aimodelsrouter.ai/docs/authentication): API keys, headers, limits and expiry.
- [Models](https://aimodelsrouter.ai/docs/models): Model slugs, the catalog endpoint and pricing fields.
- [Chat Completions](https://aimodelsrouter.ai/docs/chat-completions): Request and response schema.
- [Responses API](https://aimodelsrouter.ai/docs/responses): OpenAI Responses requests, for Codex and other Responses clients.
- [Anthropic Messages](https://aimodelsrouter.ai/docs/messages): Anthropic Messages requests, for Claude Code and the Anthropic SDKs.
- [Streaming](https://aimodelsrouter.ai/docs/streaming): Server-sent events, keepalives and the usage chunk.
- [Errors](https://aimodelsrouter.ai/docs/errors): Every error type, its HTTP status and what to do.
- [Billing & credits](https://aimodelsrouter.ai/docs/billing): Prepaid credits, holds and settlement.
- [Limits](https://aimodelsrouter.ai/docs/limits): Balance, key limits and rate limits.
- [SDKs](https://aimodelsrouter.ai/docs/sdks): OpenAI Python/TypeScript, Vercel AI SDK, LangChain.
- [FAQ](https://aimodelsrouter.ai/docs/faq): Answers to common questions.
- [OpenAPI 3.1 spec](https://aimodelsrouter.ai/openapi.json): every endpoint, parameter, response and error type

## Models

- [All models](https://aimodelsrouter.ai/models): prices, context sizes, verified capabilities and measured performance per model
- [Pricing](https://aimodelsrouter.ai/pricing): how requests and credit purchases are priced
- [DeepSeek V4.1 Flash 1017](https://aimodelsrouter.ai/models/deepseek/deepseek-v4.1-flash-1017): `deepseek/deepseek-v4.1-flash-1017`. $0.375 in / $1.50 out per 1M tokens, cached input $0.0075; 1.05M context
- [Beam 1.0 501B A23B](https://aimodelsrouter.ai/models/reflection/beam-1.0-501b-a23b): `reflection/beam-1.0-501b-a23b`. $0.50 in / $1.50 out per 1M tokens, cached input $0.05; 1.05M context
- [Nemotron 3.5 Super VL](https://aimodelsrouter.ai/models/nvidia/nemotron-3.5-super-vl): `nvidia/nemotron-3.5-super-vl`. $0.375 in / $0.8125 out per 1M tokens, cached input $0.075; 1.05M context
- [DeepSeek V4.1 Flash](https://aimodelsrouter.ai/models/deepseek/deepseek-v4.1-flash): `deepseek/deepseek-v4.1-flash`. $0.375 in / $1.50 out per 1M tokens, cached input $0.0075; 1.05M context
- [GLM 5.3 Flash](https://aimodelsrouter.ai/models/z-ai/glm-5.3-flash): `z-ai/glm-5.3-flash`. $0.1875 in / $0.625 out per 1M tokens, cached input $0.0375; 1.05M context
- [GLM 5.3](https://aimodelsrouter.ai/models/z-ai/glm-5.3): `z-ai/glm-5.3`. $1.75 in / $5.50 out per 1M tokens, cached input $0.325; 1.05M context
- [Qwen3.8 27B](https://aimodelsrouter.ai/models/qwen/qwen3.8-27b): `qwen/qwen3.8-27b`. $0.5625 in / $3.75 out per 1M tokens; 262K context
- [DeepSeek V4 Pro 0813](https://aimodelsrouter.ai/models/deepseek/deepseek-v4-pro-0813): `deepseek/deepseek-v4-pro-0813`. $1.65 in / $4.95 out per 1M tokens, cached input $0.055; 1.05M context
- [Qwen3.8 Max (open weights, 2.4T A95B)](https://aimodelsrouter.ai/models/qwen/qwen3.8-2.4t-a95b): `qwen/qwen3.8-2.4t-a95b`. $2.50 in / $7.50 out per 1M tokens, cached input $0.25; 1M context
- [DeepSeek V4 Flash 0731](https://aimodelsrouter.ai/models/deepseek/deepseek-v4-flash-0731): `deepseek/deepseek-v4-flash-0731`. $0.175 in / $0.35 out per 1M tokens, cached input $0.035; 1.05M context
- [Kimi K3](https://aimodelsrouter.ai/models/moonshotai/kimi-k3): `moonshotai/kimi-k3`. $3.75 in / $18.75 out per 1M tokens, cached input $0.375; 1.05M context; available after your first credit purchase
- [GLM 5.2](https://aimodelsrouter.ai/models/z-ai/glm-5.2): `z-ai/glm-5.2`. $1.75 in / $5.50 out per 1M tokens, cached input $0.2625; 262K context
- [Kimi K2.7 Code](https://aimodelsrouter.ai/models/moonshotai/kimi-k2.7-code): `moonshotai/kimi-k2.7-code`. $1.1875 in / $5.00 out per 1M tokens; 262K context
- [Nemotron 3 Ultra](https://aimodelsrouter.ai/models/nvidia/nemotron-3-ultra-550b-a55b): `nvidia/nemotron-3-ultra-550b-a55b`. $1.125 in / $2.125 out per 1M tokens, cached input $0.225; 131K context
- [DeepSeek V4 Flash 0423](https://aimodelsrouter.ai/models/deepseek/deepseek-v4-flash): `deepseek/deepseek-v4-flash`. $0.175 in / $0.35 out per 1M tokens, cached input $0.035; 1.05M context
- [DeepSeek V4 Pro 0423](https://aimodelsrouter.ai/models/deepseek/deepseek-v4-pro): `deepseek/deepseek-v4-pro`. $2.175 in / $4.35 out per 1M tokens, cached input $0.435; 1.05M context
- [MiMo-V2.5-Pro](https://aimodelsrouter.ai/models/xiaomi/mimo-v2.5-pro): `xiaomi/mimo-v2.5-pro`. $1.00 in / $3.75 out per 1M tokens, cached input $0.20; 262K context
- [Kimi K2.6](https://aimodelsrouter.ai/models/moonshotai/kimi-k2.6): `moonshotai/kimi-k2.6`. $1.1875 in / $5.00 out per 1M tokens, cached input $0.2375; 262K context
- [Nemotron 3 Nano Omni](https://aimodelsrouter.ai/models/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning): `nvidia/nemotron-3-nano-omni-30b-a3b-reasoning`. $0.625 in / $1.125 out per 1M tokens, cached input $0.125; 66K context
- [GLM 5.1](https://aimodelsrouter.ai/models/z-ai/glm-5.1): `z-ai/glm-5.1`. $1.75 in / $5.50 out per 1M tokens; 203K context
- [Gemma 4 31B](https://aimodelsrouter.ai/models/google/gemma-4-31b-it): `google/gemma-4-31b-it`. $0.225 in / $0.625 out per 1M tokens, cached input $0.045; 256K context
- [Qwen3.5 397B A17B](https://aimodelsrouter.ai/models/qwen/qwen3.5-397b-a17b): `qwen/qwen3.5-397b-a17b`. $0.6875 in / $4.375 out per 1M tokens, cached input $0.13875; 131K context
- [Ministral 3 14B 2512](https://aimodelsrouter.ai/models/mistralai/ministral-14b-2512): `mistralai/ministral-14b-2512`. $0.25 in / $0.25 out per 1M tokens, cached input $0.05; 262K context
- [DeepSeek V3.2](https://aimodelsrouter.ai/models/deepseek/deepseek-v3.2): `deepseek/deepseek-v3.2`. $0.625 in / $2.00 out per 1M tokens, cached input $0.1875; 164K context
- [Nemotron Nano 12B v2 VL](https://aimodelsrouter.ai/models/nvidia/nemotron-nano-12b-v2-vl): `nvidia/nemotron-nano-12b-v2-vl`. $0.25 in / $0.75 out per 1M tokens, cached input $0.05; 128K context
- [Hermes 4 405B](https://aimodelsrouter.ai/models/nousresearch/hermes-4-405b): `nousresearch/hermes-4-405b`. $1.25 in / $3.75 out per 1M tokens; 131K context
- [gpt-oss-120b](https://aimodelsrouter.ai/models/openai/gpt-oss-120b): `openai/gpt-oss-120b`. $0.125 in / $0.875 out per 1M tokens; 128K context
- [gpt-oss-20b](https://aimodelsrouter.ai/models/openai/gpt-oss-20b): `openai/gpt-oss-20b`. $0.0625 in / $0.5625 out per 1M tokens; 128K context
- [Qwen3 30B A3B Instruct 2507](https://aimodelsrouter.ai/models/qwen/qwen3-30b-a3b-instruct-2507): `qwen/qwen3-30b-a3b-instruct-2507`. $0.125 in / $0.375 out per 1M tokens; 262K context
- [Qwen3 235B A22B Instruct 2507](https://aimodelsrouter.ai/models/qwen/qwen3-235b-a22b-2507): `qwen/qwen3-235b-a22b-2507`. $0.25 in / $0.75 out per 1M tokens; 262K context
- [Llama 4 Maverick](https://aimodelsrouter.ai/models/meta-llama/llama-4-maverick): `meta-llama/llama-4-maverick`. $0.3125 in / $1.0875 out per 1M tokens, cached input $0.0625; 128K context
- [Gemma 3 27B](https://aimodelsrouter.ai/models/google/gemma-3-27b-it): `google/gemma-3-27b-it`. $0.125 in / $0.375 out per 1M tokens; 110K context

## Guides

- [Integrations](https://aimodelsrouter.ai/guides/integrations): Codex CLI, Claude Code, Cline, Kilo Code, Continue, Aider, OpenCode, Open WebUI, LibreChat, n8n, Cursor, DeepSeek Harness (DSH), OpenMausBot, LangChain, Vercel AI SDK, LlamaIndex, agent-desktop
- [Customer-support chatbot](https://aimodelsrouter.ai/guides/chatbot)
- [Extract structured data (JSON)](https://aimodelsrouter.ai/guides/extraction)
- [Read receipts, invoices and screenshots](https://aimodelsrouter.ai/guides/vision)
- [Build an agent with tool calling](https://aimodelsrouter.ai/guides/agents)
- [Classify or tag thousands of items cheaply](https://aimodelsrouter.ai/guides/classification)
- [Summarize and question long documents](https://aimodelsrouter.ai/guides/long-documents)

## Optional

- [Full reference for LLMs](https://aimodelsrouter.ai/llms-full.txt): this file plus code, parameters, errors and every model's details
- [News](https://aimodelsrouter.ai/news): new models on AI Models Router and in the industry
- [Status](https://aimodelsrouter.ai/status)
- [Terms](https://aimodelsrouter.ai/legal/terms) · [Privacy](https://aimodelsrouter.ai/legal/privacy) · Support: support@aimodelsrouter.ai