Docs

OpenAI-shaped Responses API. OpenDoor wraps it as a chat completion against the same provider resolver.

Responses

OpenAI-shaped POST /v1/responses. input is a string or a message array; OpenDoor resolves the model and calls the provider chatCompletion.

POST /v1/responses

Request Body

Field	Type	Required	Description
`model`	string	Yes	Model ID (same catalog as chat completions)
`input`	string or array	Yes	User text, or chat-style `{ role, content }` messages
`temperature`	number	No	Sampling temperature
`max_tokens`	integer	No	Maximum tokens to generate

Example

bash
curl http://localhost:3001/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.1-8b-instruct",
    "input": "What is the capital of Kenya?"
  }'

Response

json
{
  "id": "resp_abc123",
  "object": "response",
  "created_at": 1777777777,
  "model": "llama-3.1-8b-instruct",
  "status": "completed",
  "output": [
    {
      "type": "message",
      "role": "assistant",
      "content": [{ "type": "output_text", "text": "Nairobi." }]
    }
  ],
  "usage": {
    "input_tokens": 12,
    "output_tokens": 4,
    "total_tokens": 16
  }
}

Requires the same provider keys as /v1/chat/completions. Mount is documented in API_MODULES.md.

Generation

Files