Forced model via header

x-llmux-model skips cheapest-viable selection and pins the request to a catalog model, still subject to provider, capability, privacy and budget checks.

Forced model via header

examples/forced-model.json

POST /v1/chat/completions
content-type: application/json
x-llmux-model: echo-ultra

{
  "messages": [
    { "role": "user", "content": "Hello" }
  ]
}

HTTP/1.1 200 OK
x-llmux-echo: 1

{
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": "llmux demo · echo provider.\nRouted to echo/echo-ultra (tier 5, task: simple_text). No external call was made — this is a synthetic response so you can see routing and the dashboard end-to-end.\n\nYour message: \"Hello\"",
        "role": "assistant"
      }
    }
  ],
  "id": "chatcmpl-echo",
  "model": "echo-ultra",
  "object": "chat.completion",
  "usage": {
    "completion_tokens": 55,
    "prompt_tokens": 2,
    "total_tokens": 57
  }
}

override
x-llmux-model