Chat Completions

POST /chat/completions — the primary endpoint. Compatible with the OpenAI Chat Completions API.

Request

json

{
  "model": "varunsh/atlas-mini",
  "messages": [
    { "role": "system", "content": "You are concise." },
    { "role": "user", "content": "What is an API gateway?" }
  ],
  "temperature": 0.7,
  "max_tokens": 512,
  "stream": false
}

Parameters

model — required. A model id from /models.
messages — required. Array of role/content messages.
stream — optional. When true, responds with SSE.
temperature, top_p — optional sampling controls.
max_tokens — optional output cap.
tools, tool_choice — optional tool calling (model-dependent).

Response

json

{
  "id": "chatcmpl_...",
  "object": "chat.completion",
  "created": 1700000000,
  "model": "varunsh/atlas-mini",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "..." },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 18,
    "completion_tokens": 122,
    "total_tokens": 140
  }
}

Streaming

With "stream": true, the response is text/event-stream. Each event is a chat.completion.chunk; the final chunk carries usage, followed by data: [DONE].

bash

curl -N https://api-gw.aimasteryedu.in/v1/chat/completions \
  -H "Authorization: Bearer $VARUNSH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "varunsh/atlas-mini", "messages": [{"role":"user","content":"Hi"}], "stream": true }'