v1.2.1
RUN GGUF
Run LLMs locally with complete privacy. No internet required.
v1.2.1
RUN GGUF
Run LLMs locally with complete privacy. No internet required.
A powerful desktop AI chat application packed with premium features
Sleek UI with Mocha, Gruvbox, Synthwave, Matrix & Monokai themes
No internet required — everything runs locally on your machine
Blazing speed with GPU support for faster inference
Access from your phone on the same network via QR Code
Configurable reasoning depth: None, Low, Medium, High
Advanced streaming parser with state machine for reasoning tags
Animated welcome screen with beautiful typewriter effect
Background ambient sounds — Rain, White Noise & more
Export conversations in JSONL, MD, TXT, ZIP formats
Pre-configured AI personalities for different use cases
PDF, Images, and Text file support built-in
Auto-read AI responses aloud with TTS integration
Compact phone-style window for minimal footprint
Switch between llama-server and llama-cli engines
Powered by llama.cpp — running GGUF models directly on your hardware
Select a .gguf file from your device. Supports Qwen, Gemma, DeepSeek, Llama and more.
llama-server hosts the model on localhost:11434 with full GPU/CPU optimization.
Interactive interface sends and receives responses via real-time Streaming technology.
Separate HTTP server enables phone access via QR Code on your local network.
Beautiful, responsive interface designed for the best experience
Access the app from your phone on the same network
Professional state machine algorithm preventing content leak
Switch between llama-server and llama-cli
Show/hide Thinking dropdown based on settings
Windows, macOS, Linux support for llama-server
Automatic application updates
Security with rehype-sanitize
Latest version with security patches
Frontend Framework
Styling
Animations
Desktop Runtime
AI Engine
Markdown Rendering
Clone, install, and run in minutes
# Clone the repository
git clone https://github.com/YASSER-27/RunGguf.git
cd RunGguf
# Install dependencies
npm install
# Run in development mode
npm run dev
# Build production version
npm run packageDeveloper
This project was built entirely solo from scratch. Special thanks to the llama.cpp team for the amazing engine that makes running GGUF models possible on regular devices.