miniMax-M3.1-Flash

Call the MiniMax-M3.1-Flash OpenAI-compatible chat model via the baze platform. The model supports a 1,000,000-token context window with text, image, and video inputs. Thinking is always on; tune its depth with reasoning_effort. The same model is also available through POST /v1/messages and POST /v1/responses.

POST /v1/chat/completions

cURL

basic

curl --request POST \
  --url https://baze-api.powerbuyin.top/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "MiniMax-M3.1-Flash",
  "messages": [
    {
      "role": "system",
      "content": "You are a concise assistant."
    },
    {
      "role": "user",
      "content": "Describe Paris in one sentence."
    }
  ],
  "temperature": 0.3
}
'

200

{
  "id": "<string>",
  "object": "<string>",
  "created": 123,
  "model": "<string>",
  "choices": [
    {}
  ],
  "usage": {}
}

Authorizations

Authorization · string · header · required

Pass Authorization: Bearer <token> in the request header.

Body

application/json

model · enum<string> · required

Available options: MiniMax-M3.1-Flash

messages · object[] · required

Show child attributes

temperature · number · default: 0.7

Sampling temperature. Higher values make output more random.

Required range: 0 <= x <= 1

top_p · number · default: 0.95

Nucleus sampling threshold. Avoid tuning temperature and top_p at the same time.

Required range: 0 <= x <= 1

max_tokens · integer

Maximum number of output tokens per reply. The model context window is 1,000,000 tokens, covering both input and output.

Required range: x >= 1

reasoning_effort · enum<string> · default: max

Thinking depth level. MiniMax-M3.1-Flash always thinks and cannot disable it, so this field is the only way to control depth. Higher values produce more output tokens and longer waits. Defaults to max when omitted. On the Responses path, express the same setting with reasoning.effort.

Available options: low, medium, high, xhigh, max

stream · boolean

Whether to return results as an SSE event stream.

Response

200

application/json

Success. Non-streaming returns OpenAI-compatible JSON; streaming returns an OpenAI-compatible SSE event stream.

id · string

object · string

created · integer

model · string

choices · object[]

usage · object