The best provider, every request

Smart Routing & Fallback

Route each request to the best available provider by price, uptime and latency, with automatic retry and fallback when one fails.

Get startedRead the docs
Your app → LLM Gateway
Your appLLM GatewayProvider A503 · skippedProvider Bretried hereProvider Cstandby
Automatic retry on the next best provider
01

Auto routing

Price, uptime and latency weighed per request.

02

Retry & fallback

Failures move to the next provider.

03

Provider pinning

Prefix a model to lock the provider.

04

Routing savings

See what smart routing saved you.

Built for

  • High availability

    Survive provider outages without code

  • Lower spend

    Cheapest healthy provider wins

  • Per-project rules

    Tune routing per project on Enterprise

Let the gateway choose
terminal
1curl https://api.llmgateway.io/v1/chat/completions \2  -H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \3  -H "Content-Type: application/json" \4  -d '{5    "model": "gpt-4o",6    "messages": [7      {8        "role": "user",9        "content": "Hello!"10      }11    ]12  }'
Read docs

More in routing & reliability

All features

Self-Hosted or Cloud

Deploy anywhere

Anthropic-Compatible API

Point Claude Code at any model

Dynamic Routes

Routing logic, out of your code

Start using Smart Routing & Fallback today

One API key. Every model. Live in minutes.

Start buildingBrowse models