Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Setup guide - From MiniMax and Goose docs

Use MiniMax in Goose

goose's documented custom provider feature (OpenAI Compatible or Anthropic Compatible) combined with MiniMax's documented endpoints (https://api.minimax.io/v1 or https://api.minimax.io/anthropic) and model ids.

Last verified 5 steps4 official sources

Steps

  1. Step 1.

    goose Desktop: open the sidebar, click Settings, open the Models tab, click Configure providers, then click Add Custom Provider at the bottom of the window.

  2. Step 2.

    Fill in the provider details: Provider Type = OpenAI Compatible; Display Name = MiniMax; API URL = MiniMax's OpenAI-compatible base URL; API Key = your MiniMax key; Available Models = comma-separated MiniMax model names; Streaming Support on. Then click Create Provider.

    Snippet
    API URL: https://api.minimax.io/v1
    Available Models: MiniMax-M3, MiniMax-M2.7
  3. Step 3.

    goose CLI alternative: run goose configure, select Custom Providers, then Add A Custom Provider, and enter the same details when prompted.

    Terminal
    goose configure
  4. Step 4.

    Config file alternative: create a JSON file in ~/.config/goose/custom_providers/ (Windows: %APPDATA%\Block\goose\config\custom_providers\) following goose's documented schema, filled with MiniMax values.

    JSON
    {
      "name": "minimax",
      "engine": "openai",
      "display_name": "MiniMax",
      "description": "MiniMax OpenAI-compatible API",
      "api_key_env": "MINIMAX_API_KEY",
      "base_url": "https://api.minimax.io/v1/chat/completions",
      "models": [
        {
          "name": "MiniMax-M3",
          "context_limit": 1000000
        },
        {
          "name": "MiniMax-M2.7",
          "context_limit": 204800
        }
      ],
      "supports_streaming": true,
      "requires_auth": true
    }
  5. Step 5.

    Set the key named in api_key_env and start a session with the provider.

    Terminal
    export MINIMAX_API_KEY="your-api-key"
    goose session start --provider minimax

Caveats

  • Combination: goose documents the custom provider mechanism; MiniMax documents the endpoint and model ids. Neither publishes a MiniMax-specific goose guide, and goose has no native MiniMax provider in its provider table.
  • The config-file example is goose's custom_corp_api.json schema filled with MiniMax values. goose's own example base_url includes the full /v1/chat/completions path, which MiniMax documents as POST https://api.minimax.io/v1/chat/completions; the provider name and api_key_env name (MINIMAX_API_KEY) are arbitrary choices.
  • Context limits come from MiniMax's docs (1,000,000 for M3 / M3.1-Flash-Preview, 204,800 for M2.x).
  • goose also supports Anthropic Compatible custom providers; MiniMax's Anthropic-compatible base URL is https://api.minimax.io/anthropic.
  • MiniMax-M3.1-Flash-Preview requires an M Plan Subscription Key.

FAQ

How do I use MiniMax in Goose?

goose's documented custom provider feature (OpenAI Compatible or Anthropic Compatible) combined with MiniMax's documented endpoints (https://api.minimax.io/v1 or https://api.minimax.io/anthropic) and model ids.

Which MiniMax model ids work in Goose?

The official docs read for this guide list MiniMax-M3, MiniMax-M2.7, MiniMax-M2.7-highspeed, MiniMax-M2.5, MiniMax-M2.5-highspeed and MiniMax-M3.1-Flash-Preview. Requests go to https://api.minimax.io/v1.

Is this an official MiniMax and Goose setup?

The steps draw on both MiniMax's and Goose's docs. It was last checked against 4 official pages on October 17, 2026.

Sources

Last verified against 4 official pages.

Newsletter

New models, set up the day they ship

One email per week. New model releases, the coding tools that support them and how to wire them up.