PyllamaUI for VS Code 🧠🦙

A lightweight offline AI assistant inside VS Code powered by Python + Ollama.
Access local LLMs directly from your editor — no cloud, no telemetry, no internet required.

🌟 About

PyllamaUI (VS Code Edition) is a VS Code extension that lets you chat with locally hosted LLMs using Python as the backend and Ollama as the model runner.

It's fully offline, privacy-friendly, and designed for low-resource systems.

✨ Features

💬 Chat UI inside a VS Code panel
🧠 Interact with local Ollama models (llama3, mistral, etc.)
🐍 Python-powered backend (via run_prompt.py)
🔌 Uses VS Code’s WebView for integrated GUI
🚫 Works without internet – offline-first mindset
📁 All user data processed locally

🛠️ Requirements

💻 System

VS Code 1.75+
Python 3.10+
ollama

✅ Model Recommendations

Use Case	Model Name	Approx. Size	Description
📝 Text Chat	`tinyllama`	~600 MB	Lightweight text model
💻 Coding Help	`deepseek-coder`	~700 MB	Designed for code generation
⚡ All-in-One	`gemma:3b` (aka `gemma3n`)	~5.5 GB	Great for both chat & coding (Google DeepMind)

✅ If unsure, just install gemma3n for the best all-around experience.

To run a model:
ollama run gemma3n
ollama run tinyllama
ollama run deepseek-coder

📌 Special thanks to Ollama for making local LLMs accessible to all.

Developed with ❤️ by BHUVANESH M