Kimi AI for Copilot Chat

Pick Kimi K2.5 & K2.6 from the Copilot Chat model picker — and keep everything else Copilot already gives you.

Love Kimi's price-performance but don't want to give up GitHub Copilot's agent mode, tool calling, and polished UI? This extension drops Kimi K2.5 & K2.6 straight into the Copilot Chat model selector — with vision, video support, thinking mode, and your own API key.

Why this extension?

• Don't replace Copilot — power it up. No new sidebar, no new chat UI to learn. Just a new model in the picker you already use.
• Agent mode, tool calling, instructions, MCP, skills — all of it still works. Copilot's entire stack, now running on Kimi.
• Vision on a text-only model. Kimi K2.x natively supports images and video, so attachments work directly. The vision proxy is available as a fallback.
• BYOK, pay Moonshot directly. Your API key, your bill, your rate limits. Stored in the OS keychain, never on disk.

Features

Kimi K2.5 & K2.6 in the model picker

Both models show up alongside GPT-4o, Claude, and friends in Copilot Chat's model selector. 262K token context on both. Switch models mid-chat without losing history.

Vision & Video Support

Kimi K2.x models natively support images and video input. Drop a screenshot or video frame into chat and Kimi can see it directly.

Thinking Mode with Reasoning Effort Control

Full support for Kimi's reasoning capabilities. Use Copilot Chat's native model picker menu to choose none (off), high (balanced, default), or max (deep reasoning for hard agent tasks).

Inherits Every Copilot Capability

Because this plugs into Copilot's native provider API, you get the full stack for free:

• Agent mode — autonomous multi-step tasks
• Tool calling — file edits, terminal, workspace search, Git, tests
• Instructions & skills — all your .instructions.md, AGENTS.md, and skills just work

Secure by Default

API key lives in VS Code's SecretStorage (OS keychain on macOS / Windows / Linux). Never in settings.json, never in your Git history.

Zero Runtime Dependencies

Pure VS Code API + Node.js built-ins. No Python, no Docker, no local proxy server to babysit.

Getting Started

Prerequisites

• VS Code 1.116 or later
• GitHub Copilot subscription (Free / Pro / Enterprise — the free tier works)
• Kimi API key from platform.moonshot.ai, or a compatible provider token when using a custom kimi-copilot.baseUrl

Usage

1. Install from the VS Code Marketplace (or load the .vsix)
2. Run Kimi: Set API Key from the Command Palette (Cmd+Shift+P)
3. Paste your Moonshot API key (usually starts with sk-)
4. Open Copilot Chat, click the model picker, pick Kimi K2.5 or Kimi K2.6
5. That's it — chat away

Models

Both support thinking mode, tool calling, and 262K token context.

Settings

Thinking Effort is configured from Copilot Chat's model picker for each Kimi model.

Example settings.json override for compatible API proxies:

{
  "kimi-copilot.modelIdOverrides": {
    "kimi-k2.5": "your-k2.5-model-id",
    "kimi-k2.6": "your-k2.6-model-id"
  }
}

Compared to alternatives

Credits

This extension is based on DeepSeek V4 for Copilot Chat by Vizards, adapted for the Kimi (Moonshot AI) API. All credit for the original extension architecture goes to the Vizards team.

License

MIT