API

Chat Completions

Request format, response format, and current behavior of the OpenAI-compatible endpoint.

Public base URL

Routera publishes the API at https://routera.one/api/v1.

Routera’s main generation endpoint is POST /chat/completions.

Endpoint

POST https://routera.one/api/v1/chat/completions

Request body

Routera currently reads these fields:

Field	Required	Notes
`model`	Yes	Must match a model id from `GET /models`
`messages`	Yes	Array of chat messages
`stream`	No	When `true`, returns SSE chunks

Example:

{
  "model": "openai/gpt-5.5",
  "messages": [
    { "role": "system", "content": "Be concise." },
    { "role": "user", "content": "Summarize Routera in one sentence." }
  ]
}

Response body

Non-streaming responses use an OpenAI-style shape:

{
  "id": "chatcmpl_x",
  "object": "chat.completion",
  "created": 1710000000,
  "model": "openai/gpt-5.5",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Routera is an OpenAI-compatible routing layer for model access."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 17,
    "total_tokens": 29
  }
}

Current compatibility

Routera is intentionally narrow right now.

The endpoint is OpenAI-compatible in shape.
The route currently documents only model, messages, and stream as supported request fields.
Unknown OpenAI fields should not be relied on yet unless we explicitly add support for them.