Skip to content
| Marketplace
Sign in
Visual Studio Code>Programming Languages>AI Chatbot — Local & Custom ModelsNew to Visual Studio Code? Get it now.
AI Chatbot — Local & Custom Models

AI Chatbot — Local & Custom Models

Akash Maddheshiya

|
3 installs
| (0) | Free
AI chat sidebar for VS Code powered by your own local model (Ollama, LM Studio) or any OpenAI-compatible backend. Chat, explain code, review code, generate code, inline completions, conversation history.
Installation
Launch VS Code Quick Open (Ctrl+P), paste the following command, and press enter.
Copied to clipboard
More Info

AI Chatbot VS Code Extension

A VS Code extension that connects to any local or custom AI model backend.

Features

Feature How to use
Chat sidebar Click the chat bubble icon in the Activity Bar
Explain code Select code → right-click → AI: Explain Selected Code (or Ctrl+Shift+E)
Review code Select code → right-click → AI: Review Selected Code
Generate code Right-click → AI: Generate Code from Prompt (or Ctrl+Shift+G)
Inline completions Just type — press Tab to accept suggestions

Setup

1. Install dependencies

cd chatbot-ide-plugin
npm install
npm run compile

2. Run in development

Open the folder in VS Code, press F5 — a new Extension Development Host window opens.

3. Package as .vsix

npm run package
# installs as: code --install-extension chatbot-ide-plugin-0.1.0.vsix

4. Configure your model

Open Settings → Extensions → AI Chatbot and set:

Setting Default Description
chatbot.modelEndpoint http://localhost:11434/api/chat Your model API URL
chatbot.modelName llama3 Model name
chatbot.apiFormat ollama ollama, openai-compatible, or custom
chatbot.apiKey (blank) API key if required
chatbot.systemPrompt coding assistant System prompt
chatbot.inlineSuggestEnabled true Toggle inline suggestions

Backend examples

Ollama (local)

ollama serve          # starts at http://localhost:11434
ollama pull llama3    # download a model

Set apiFormat = ollama, endpoint = http://localhost:11434/api/chat, modelName = llama3.

LM Studio

Start the local server in LM Studio (OpenAI-compatible).
Set apiFormat = openai-compatible, endpoint = http://localhost:1234/v1/chat/completions.

Custom backend

Your backend must accept POST with { model, messages } and return { response: "..." }.
Set apiFormat = custom.

  • Contact us
  • Jobs
  • Privacy
  • Manage cookies
  • Terms of use
  • Trademarks
  • Your Privacy Choices
  • Consumer Health Privacy
© 2026 Microsoft