Chat Completions

The /chat/completions endpoint is OpenAI-compatible. If your code already talks to OpenAI chat completions, it works here unchanged.

Request

Common fields:

• model — a served model id (see Models).
• messages — array of {role, content} objects (system, user, assistant).
• max_tokens — max tokens to generate.
• temperature, top_p, stop — standard sampling controls.
• stream — set true for token streaming (see Streaming).

POST https://api.jouledns.com/v1/chat/completions
Authorization: Bearer <your-org-key>
Content-Type: application/json

{
  "model": "qwen2.5-7b-instruct",
  "messages": [
    {"role": "system", "content": "You are concise."},
    {"role": "user", "content": "What is an inference router?"}
  ],
  "max_tokens": 256,
  "temperature": 0.7
}

Response

A standard chat-completion object. The generated text is at choices[0].message.content; token accounting is under usage.

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "qwen2.5-7b-instruct",
  "choices": [
    {
      "index": 0,
      "message": {"role": "assistant", "content": "..."},
      "finish_reason": "stop"
    }
  ],
  "usage": {"prompt_tokens": 24, "completion_tokens": 40, "total_tokens": 64}
}