Quick start

Drop a config, start the server, call it like OpenAI.

1 · config & run

config & run

cp models.yaml ~/.config/routsi/   # or keep ./models.yaml
routsi serve                       # listens on :8080
routsi serve -port 9000            # …or override the port

2 · call it (any OpenAI SDK works)

curl

curl localhost:8080/v1/chat/completions \
  -H 'content-type: application/json' \
  -d '{"model":"auto","messages":[{"role":"user","content":"hello"}]}'

The chosen answerer is returned in the X-Selected-Model response header, and every response carries a usage block.

Next: Core concepts →