MiniMax icon

MiniMax

Consume MiniMax AI API (Intl + China, Standard + Coding Plan)

Chat → Complete

AI-generated

Summary

Generate a chat completion response from a MiniMax model for a given conversation prompt, returning either a simplified result or the raw API data.

Inputs

  • Model (required) — Select a MiniMax chat model from the list or specify a custom model ID supported by your MiniMax account.
  • Prompt (required) — Provide one or more chat messages as the prompt, each with a role (system, user, or assistant) and content text.
  • Simplify — Optionally simplify the API response to return only the assistant's message content instead of the full raw data.
  • Options — Optional parameters to control generation such as Frequency Penalty, Maximum Number of Tokens, Presence Penalty, Response Format (text or JSON object), Sampling Temperature, and Top P sampling parameter. These are applied to the request body to influence the model output.

Output shape

a single item containing the chat completion response message or messages

If simplification is enabled, the output contains the assistant's response message directly. Otherwise, the complete raw data from the MiniMax API is returned. Multiple messages may be included if multiple completion choices are returned.

Links

Discussion