🤖 Local AI Pilot

Supercharge your coding with Local AI models!

A lightweight extension unlocking powerful AI models that run directly on your machine, keeping your code secure, private and responsive.

⛓ Key Features:

AI-powered assistance: Get real-time code completion, chat with the AI about your code, and tackle complex tasks.

Local Ollama models: Leverage the power of Ollama for a smooth offline experience and complete control over your data.

Fully customizable: Use containers to tailor the extension to your specific needs and preferences.

Completion with Context: Get suggestions tailored to your code's specific situation.

Document Q&A: Ingest your documents and ask questions offline using RAG (Retrieval-Augmented Generation)

Chat History: Keep track of your conversations and utilize past interactions for future reference.

Remote Models: Supports remote inference using providers like Open AI, Gemini, Cohere, Anthropic, Codestral for resource-constrained devices.

🚀 Quick Start

Local setup

The extension supports Standalone Mode (Default) / Container Mode

Configurable in Settings > Extensions > Local AI Pilot > Mode

Standalone Mode - The extension connects directly with Ollama container/standalone instance.

• Standalone Mode Configuration

Container Mode - An intermediate API service bridges Ollama service with the extension allowing extra configurations/features.

• Container Mode Configuration

📖 Usage

💭 Code chat

📝 Code completion

Note: Use LF line endings for proper formatting.

📝 Code completion with context

Provide the paths of files to use as additional context during code completion.

🛠️ Explain, Fix or Review code blocks

📃 Document Q & A (container mode)

📜 Chat History & Caching (container mode)

Credits

• Model inference - Ollama llamaindex
• Styling - tailwindcss prism flowbite
• Code icons created by Azland Studio - Flaticon