AI API ROUTER
One API.
Every AI model.
9Router routes your requests to the best AI model for the job — with automatic fallbacks, cost optimization, and a single OpenAI-compatible endpoint. Stop juggling API keys.
curl https://api.9router.dev/v1/chat/completions \ -H "Authorization: Bearer $ROUTER_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "9router/auto", "messages": [{"role": "user", "content": "Summarize this thread"}] }'import requests r = requests.post( "https://api.9router.dev/v1/chat/completions", headers={"Authorization": "Bearer $ROUTER_KEY"}, json={ "model": "9router/auto", # smart routing "messages": [{"role": "user", "content": "Summarize this thread"}], }, ) print(r.json()["choices"][0]["message"]["content"])const res = await fetch("https://api.9router.dev/v1/chat/completions", { method: "POST", headers: { "Authorization": `Bearer ${process.env.ROUTER_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ model: "9router/auto", // smart routing messages: [{ role: "user", content: "Summarize this thread" }], }), }); const data = await res.json(); console.log(data.choices[0].message.content);
Drop-in compatible with the OpenAI API — change the base URL and you're live.
ROUTES ACROSS LEADING PROVIDERS
FEATURES
Infrastructure for the multi-model era
One integration. Every frontier model, routed intelligently.
Smart routing
Route by cost, latency, or quality. Use 9router/auto and let the router pick the best model per request — or pin a specific model anytime.
Automatic fallbacks
Provider outage? Rate limit? Requests fail over to your next-best model instantly, with zero code changes on your side.
Cost optimization
Route simple tasks to fast, cheap models and reserve frontier models for hard problems. Cut inference spend without touching your code.
OpenAI-compatible
Works with the SDKs and tools you already use. Swap the base URL and API key — chat, embeddings, and structured outputs just work.
Observability
Every request logged: model used, tokens, latency, and cost. Filter by key, user, or app and see exactly where your budget goes.
One key, unified billing
A single API key and one invoice for every provider. Set per-key budgets and rate limits for your team or your customers.
HOW IT WORKS
Live in three steps
Get an API key
Sign up and generate a key in seconds. No credit card required to start.
Point at api.9router.dev
Change your base URL to https://api.9router.dev/v1. Your existing OpenAI client keeps working.
Set your routing policy
Choose 9router/auto for smart routing, or pin models and fallbacks per request.
MODELS
Access the models that matter
A sample of what's routable today — new models are added as providers release them.
| Model | Provider | Best for | Context |
|---|---|---|---|
anthropic/claude-opus-4-6 | Anthropic | Complex reasoning, coding agents | 200K |
anthropic/claude-sonnet-4-6 | Anthropic | Balanced quality & speed | 200K |
openai/gpt-5 | OpenAI | General purpose, tools | 400K |
google/gemini-2.5-pro | Long context, multimodal | 1M | |
xai/grok-4 | xAI | Real-time knowledge | 256K |
deepseek/deepseek-v3 | DeepSeek | Cost-efficient coding | 128K |
Model availability follows each provider's terms. Router adds a small, transparent margin on provider pricing.
PRICING
Start free. Scale when you do.
MOST POPULAR
Pro
$20/mo
- $25 in monthly credits
- Higher rate limits
- Priority fallbacks
- Email support
ABOUT
Built for developers, by developers
9Router is an independent AI infrastructure project founded in 2026. We believe the future is multi-model: no single provider wins every workload, so developers need a neutral layer that routes each request to the right model — reliably and affordably.
Stop juggling API keys.
Get one key, route to every model, and ship faster.
Get your API key