✦ Now supporting Claude, ChatGPT, Grok, Gemini & more

Deploy AI Agents as API Endpoints

Create custom AI agents with any provider. Get a unique endpoint. Start sending requests in seconds.

Create Free Agent → Read the Docs
quickstart.sh
# 1. Create your agent
curl -X POST https://airouter.us/api/v1/agents \
  -H "Authorization: Bearer $AIROUTER_KEY" \
  -d '{
    "name": "mechagrok",
    "provider": "xai",
    "model": "grok-3",
    "system_prompt": "You are a sharp Twitter debater..."
  }'

# → {"agent_id": "nc38ybrf", "endpoint": "nc38ybrf.xai.airouter.us"}

# 2. Use your agent
curl https://nc38ybrf.xai.airouter.us/v1/chat/completions \
  -H "Authorization: Bearer $AIROUTER_KEY" \
  -d '{"messages": [{"role": "user", "content": "Roast this take..."}]}'
Supported Providers
Anthropic
OpenAI
xAI
Google
Flux
Stable Diffusion

Built for developers

Everything you need to deploy, manage, and scale AI agents through a unified API.

Unique Endpoints

Each agent gets a dedicated subdomain. Route requests to any provider through a single, consistent API format.

🔑

Bring Your Own Keys

Use your own API keys from any provider. We never store or access your data — pure passthrough with your instructions baked in.

🌊

Streaming & SSE

Full support for streaming responses via Server-Sent Events. Real-time token delivery with sub-100ms first-byte latency.

🖼️

Image Generation

Generate images through Flux, DALL-E, and Stable Diffusion. Same endpoint pattern, same auth, same simplicity.

📊

Usage Analytics

Track tokens, requests, latency, and costs per agent. Export logs for debugging or billing reconciliation.

🛡️

Rate Limiting & Guards

Set per-agent rate limits, content filters, and budget caps. Prevent runaway costs with automatic circuit breakers.

Transparent pricing

Pay provider rates + a small platform fee. No hidden costs, no minimums.

Provider / Model Input Output Platform Fee
Anthropic / Claude $3.00/M $15.00/M +5%
OpenAI / ChatGPT $2.00/M $8.00/M +5%
xAI / Grok $3.00/M $15.00/M +5%
Google / Gemini $1.25/M $10.00/M +5%
Flux / Schnell $0.003/image +5%

Three steps to deploy

From zero to a production AI agent endpoint in under a minute.

01

Pick a provider

Choose from Anthropic, OpenAI, xAI, Google, or image generation models.

02

Configure your agent

Set a name, system prompt, model parameters, and paste your provider API key.

03

Get your endpoint

Receive a unique subdomain. Start sending requests immediately — OpenAI-compatible format.