Skip to content

Ollama Local Chat

VS Code extension to chat with a local Ollama instance inside the editor—similar to Copilot-style chat but using your own models.

Requirements

Ollama installed and running locally
VS Code 1.74.0 or newer

Installation

Install dependencies: npm install
Build: npm run compile
Press F5 in VS Code to run the extension in a new window (Extension Development Host)

To install from source in your main VS Code:

Run npm run compile
In VS Code: Run → Install Additional Development Extensions... → choose Install from VSIX... and pick the built VSIX, or run the extension from this folder via Run and Debug (F5).

Usage

Start Ollama (e.g. ollama serve or ensure the Ollama app is running).
Open the Command Palette: Cmd+Shift+P (macOS) or Ctrl+Shift+P (Windows/Linux).
Run Open Ollama Chat.
Use the chat panel to send messages; responses stream in real time.

Configuration

In VS Code Settings (or settings.json):

Setting	Description	Default
`ollama.baseUrl`	Ollama API base URL	`http://localhost:11434`
`ollama.model`	Default model for the chat	`llama3.2:latest`

Features

Chat panel inside VS Code
Connects to your local Ollama API
Streaming responses
Theme-aware UI
Configurable model and base URL

License

MIT