Skip to main content

MiniMax

MiniMax serves the MiniMax family of LLMs (for example MiniMax-M3) through an OpenAI-compatible API.

Getting started with MiniMax

  1. Sign in to platform.minimax.io.
  2. Create an API key from your account.
  3. Pick a model from the MiniMax model docs (for example MiniMax-M3).

Setting up MiniMax in Unstract

  • Sign in to the Unstract Platform.
  • From the side navigation menu, choose Settings 🞂 LLMs.
  • Click on the New LLM Profile button.
  • From the list of LLMs, choose MiniMax.
  • Name — a unique name for this adapter instance.
  • API Key — your MiniMax API key from platform.minimax.io.
  • Model — the model name from the MiniMax docs. Defaults to MiniMax-M3.
  • API Base — pre-filled with https://api.minimax.io/v1. A base whose path ends in /anthropic uses the Anthropic-compatible protocol; any other base uses the OpenAI-compatible protocol. China accounts use the api.minimaxi.com host.
  • Service Tierstandard (default) or priority (faster admission at a higher price). Cost tracking records standard-tier rates.
  • Maximum Output Tokens — defaults to 4096. Leave empty to use the provider default.
  • Enable Thinking — overrides the protocol default for MiniMax-M3. MiniMax-M2.x models always keep thinking enabled.
  • Leave Max Retries and Timeout at their defaults.
  • Click on Test Connection and ensure it succeeds, then click Submit.