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

Setup guide - From MiniMax and GitHub Copilot docs

Use MiniMax in GitHub Copilot

VS Code's Copilot Chat 'Custom Endpoint' BYOK provider (Chat Completions API type) combined with MiniMax's documented Chat Completions endpoint https://api.minimax.io/v1/chat/completions and model ids.

Last verified 4 steps4 official sources

Steps

  1. Step 1.

    Open the Language Models editor: select Manage Language Models (gear icon) from the model picker, or run Chat: Manage Language Models from the Command Palette.

  2. Step 2.

    Select Add Models, then select Custom Endpoint. Enter a group name, then a display name and your MiniMax API key.

  3. Step 3.

    Select the API type Chat Completions. VS Code opens chatLanguageModels.json; configure the model using MiniMax's endpoint and model id, then save.

    JSON
    [
  {
    "name": "MiniMax",
    "vendor": "customendpoint",
    "apiKey": "${input:minimaxApiKey}",
    "apiType": "chat-completions",
    "models": [
      {
        "id": "MiniMax-M3",
        "name": "MiniMax-M3",
        "url": "https://api.minimax.io/v1/chat/completions",
        "toolCalling": true,
        "vision": true,
        "contextWindow": 1000000,
        "maxOutputTokens": 64000
      }
    ]
  }
]
  4. Step 4.

    Select the model from the model picker in chat. If it does not appear immediately, restart VS Code.

Caveats

  • Combination: VS Code documents the Custom Endpoint mechanism; MiniMax documents POST https://api.minimax.io/v1/chat/completions with bearer auth plus the model ids. Neither publishes a MiniMax-specific Copilot guide. This applies to Copilot Chat in VS Code, not Copilot on github.com or other IDEs.
  • The chatLanguageModels.json snippet is VS Code's documented schema filled with MiniMax values; maxOutputTokens 64000 is an illustrative choice, not a MiniMax-published limit. vision is true only for M3 / M3.1-Flash-Preview (M2.x models do not accept image input).
  • For the Chat Completions type, VS Code sends Authorization: Bearer <apiKey> by default, which matches MiniMax's bearer auth.
  • For MiniMax-M3.1-Flash-Preview, VS Code's optional thinking / supportsReasoningEffort ["low","medium","high","xhigh","max"] / reasoningEffortFormat "chat-completions" fields map to MiniMax's reasoning_effort parameter; never send none (MiniMax returns 400). M3.1-Flash-Preview needs an M Plan Subscription Key.
  • Copilot Business/Enterprise users need their admin to enable the 'Bring Your Own Language Model Key in VS Code' policy. BYOK applies to chat and utility tasks only; inline suggestions, semantic search and embeddings still use GitHub's models.
  • The github.copilot.chat.customOAIModels setting and the OpenAI Compatible provider are deprecated in favor of Custom Endpoint.

More setup guides

Related reading

FAQ

How do I use MiniMax in GitHub Copilot?

VS Code's Copilot Chat 'Custom Endpoint' BYOK provider (Chat Completions API type) combined with MiniMax's documented Chat Completions endpoint https://api.minimax.io/v1/chat/completions and model ids.

Which MiniMax model ids work in GitHub Copilot?

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

Is this an official MiniMax and GitHub Copilot setup?

The steps draw on both MiniMax's and GitHub Copilot'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.