AnRouter
en-US zh-CN Get started

One API for China’s Frontier AI Models
Leading open-source ecosystems, one unified interface, flexible access.

Access DeepSeek, GLM, MiniMax, Qwen, and MiMo through OpenAI-compatible APIs to build production AI agents and automated workflows at lower cost.

POST/v1/chat/completions
Request
curl -X POST "https://api.anrouter.com/v1/chat/completions" \
  -H "Authorization: Bearer sk-••••" \
  -d '{
    "model": "glm-5.3",
    "messages": [
      { "role": "user", "content": "..." }
    ]
  }'

Extended context, no chunking

Context windows up to one million tokens. Complete codebases, long-form legal documents, and multi-source research can be processed in a single API call.

Built for production economics

Frontier reasoning at rates designed for sustained, high-volume usage. Committed usage unlocks volume discounts across every model.

Mainstream protocols, directly compatible

OpenAI and Anthropic protocols are directly supported without rewriting existing application logic. Point any compatible SDK at AnRouter and it works.

Works with the tools developers already use

OpenCode, Codex, Cline, Claude Code, and every major AI development tool integrates with AnRouter through standard protocols. No application code changes and no prompt rewrites.

Cursor
Windsurf
GitHub Copilot
Claude Code
VS Code
OpenClaw
OpenCode
Cline
LangChain
Dify
n8n
OpenAI SDK
Ollama
Hugging Face
Vercel AI
Perplexity
GitHub
Qoder
Codex CLI
OpenRouter

The model lineup

Eleven models cover reasoning, coding, multimodal input, long context, and agent workflows through one endpoint with consistent API semantics. Card prices are input, cached input, and output.

Explore All Models →

DeepSeek

deepseek-v4-flash-0731

Multimodal · Coding
Input / cache / output · per 1M tokens Available
Default$0.03 / $0.012 / $0.14
Pro · 20% off$0.024 / $0.012 / $0.112
Max · 50% off$0.015 / $0.012 / $0.07
Context1M

DeepSeek

deepseek-v4.1-flash

Multimodal · Reasoning
Input / cache / output · per 1M tokens Available
Default$0.08 / $0.015 / $0.3
Pro · 20% off$0.064 / $0.015 / $0.24
Max · 50% off$0.04 / $0.015 / $0.15
Context1M

Z.AI

glm-5

Multimodal · Reasoning
Input / cache / output · per 1M tokens Available
Default$0.25 / $0.05 / $0.8
Pro · 20% off$0.2 / $0.05 / $0.64
Max · 50% off$0.125 / $0.05 / $0.4
Context202,752

Z.AI

glm-5.2

Multimodal · Reasoning
Input / cache / output · per 1M tokens Available
Default$0.42 / $0.08 / $1.25
Pro · 20% off$0.336 / $0.08 / $1
Max · 50% off$0.21 / $0.08 / $0.625
Context1M

Z.AI

glm-5.3

Multimodal · Coding
Input / cache / output · per 1M tokens Available
Default$0.45 / $0.08 / $1.3
Pro · 20% off$0.36 / $0.08 / $1.04
Max · 50% off$0.225 / $0.08 / $0.65
Context1M

Z.AI

glm-5.3-flash

Multimodal · Fast
Input / cache / output · per 1M tokens Available
Default$0.045 / $0.012 / $0.16
Pro · 20% off$0.036 / $0.012 / $0.128
Max · 50% off$0.0225 / $0.012 / $0.08
Context1M

MiniMax

minimax-m3

Text · Reasoning
Input / cache / output · per 1M tokens Available
Default$0.12 / $0.038 / $0.4
Pro · 20% off$0.096 / $0.038 / $0.32
Max · 50% off$0.06 / $0.038 / $0.2
Context524K

Qwen

qwen3.8-27b

Multimodal · Reasoning
Input / cache / output · per 1M tokens Available
Default$0.08 / $0.025 / $0.4
Pro · 20% off$0.064 / $0.025 / $0.32
Max · 50% off$0.04 / $0.025 / $0.2
Context262K

Qwen

qwen3.8-flash-next

Multimodal · Fast
Input / cache / output · per 1M tokens Available
Default$0.06 / $0.014 / $0.12
Pro · 20% off$0.048 / $0.014 / $0.096
Max · 50% off$0.03 / $0.014 / $0.06
Context262K

Xiaomi MiMo

mimo-v2.6-pro

Multimodal · Coding
Input / cache / output · per 1M tokens Available
Default$0.25 / $0.04 / $0.5
Pro · 20% off$0.2 / $0.04 / $0.4
Max · 50% off$0.125 / $0.04 / $0.25
Context1M

Xiaomi MiMo

mimo-v2.6-flash

Multimodal · Fast
Input / cache / output · per 1M tokens Available
Default$0.06 / $0.01 / $0.15
Pro · 20% off$0.048 / $0.01 / $0.12
Max · 50% off$0.03 / $0.01 / $0.075
Context1M

Get started

Start building with AnRouter.

Free API access. No credit card required.

Contact Sales Get Free API Key →