cURL

Chat completion

curl https://api.routeur.ai/v1/chat/completions \
  -H "Authorization: Bearer $ROUTEUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [{"role": "user", "content": "hi"}]
  }'

Streaming

Add "stream": true to receive server-sent events. Use curl -N to disable buffering:

curl -N https://api.routeur.ai/v1/chat/completions \
  -H "Authorization: Bearer $ROUTEUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "stream": true,
    "messages": [{"role": "user", "content": "stream please"}]
  }'

Choosing a model

Every request picks an upstream model in one of three ways — in this order of precedence:

1. Routeur-* request headers. If Routeur-Provider or Routeur-Model are set, they win, regardless of what is in the body. Useful from middleware or transports that can't change the JSON.
2. An explicit upstream id in model. Pass something like gpt-4o-mini or claude-3-5-sonnet to bypass routing rules and pin the request to that model.
3. model: "auto" (or any alias). Lets routeur.ai pick the upstream model from your routing rules. This is the recommended default.

Force a specific provider+model on a single call via headers (the body still needs a model field — it is ignored when headers are set):

curl https://api.routeur.ai/v1/chat/completions \
  -H "Authorization: Bearer $ROUTEUR_KEY" \
  -H "Routeur-Provider: openai" \
  -H "Routeur-Model: gpt-4o" \
  -H "Content-Type: application/json" \
  -d '{ "model": "auto", "messages": [{"role":"user","content":"hi"}] }'

Production tips

• Always set a connect + read timeout on your HTTP client. The platform default is rarely what you want.
• Stream long completions to surface first-token latency and let users cancel.
• Persist the X-Routeur-Trace-* response headers if you want to inspect a routed request after the fact — see Inspect trace.