Legacy

Legacy /v1/completions

The original completions endpoint, kept for backward compatibility with text-davinci-003 callers. Chat completions is the modern, recommended path.

⚠️

Migrate to /v1/chat/completions

The legacy completions endpoint is frozen and will not receive new features, model upgrades, or performance improvements. All active development targets the chat completions API. If you are still sending prompts to /v1/completions, plan your migration now.

View chat completions docs

Endpoint

POST https://api.getnimbus.net/v1/completions

Request Body

{
  "model": "text-davinci-003",
  "prompt": "Say hello to Meridian.",
  "max_tokens": 64,
  "temperature": 0.7,
  "top_p": 1,
  "n": 1,
  "stream": false,
  "stop": null
}

Only model text-davinci-003 is available. All other parameters mirror the OpenAI legacy completions spec.

Response

{
  "id": "cmpl-abc123",
  "object": "text_completion",
  "created": 1712600000,
  "model": "text-davinci-003",
  "choices": [
    {
      "text": " Hello! How can I assist you today?",
      "index": 0,
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 5,
    "completion_tokens": 8,
    "total_tokens": 13
  }
}

Rate Limits

Free tier

3 requests / minute

Pro tier

60 requests / minute

← Chat completionsModels →