Run local LLMs without the hassle
Features • Install • Build • Binaries • Contribute
A beautiful, modern desktop application for managing and interacting with local AI models. Run AI entirely on your machine — no data leaves your computer.
┌─────────────────────────────────────┐
│ INSTALLER SIZE ████████ 5MB│
│ GPU ACCELERATION ████████ YES │
│ PRIVACY ████████ 100%│
│ OFFLINE ████████ YES │
└─────────────────────────────────────┘
- Tiny installer — 5MB download, binaries fetched on-demand
- GPU acceleration — Auto-detect and download CUDA/Vulkan/Metal binaries
- 100% private — Everything runs locally, no data leaves your machine
- HuggingFace integration — Search and download GGUF models directly
- AI-powered recommendations — Get model suggestions based on your hardware
- Clean, dark-themed interface with smooth animations
- Intuitive navigation and responsive design
- Real-time progress indicators for downloads
- My Library: View and manage your downloaded models
- Browse: Discover curated AI models optimized for local execution
- HuggingFace Integration: Search and download GGUF models from HuggingFace
- Model Wizard: AI-powered recommendations based on your hardware
- Multi-conversation support
- Real-time streaming responses
- Model switching on the fly
- Conversation history
- Automatic retry on network failures (up to 3 attempts)
- Resume capability for interrupted downloads
- Real-time progress tracking
- Intelligent error handling
- Hardware detection (GPU, RAM, CPU)
- On-demand binary downloads for GPU acceleration
- Optimization settings
- License management (Pro features)
- Performance dashboard
Download the latest release for your platform:
| Platform | Download | Size |
|---|---|---|
| Windows | OpenLLM-Studio_x64-setup.exe |
~5 MB |
| macOS (Apple Silicon) | Coming Soon | - |
| macOS (Intel) | Coming Soon | - |
| Linux | Coming Soon | - |
On first launch, the app will:
- Detect your hardware — GPU, RAM, CPU
- Recommend the best runtime — CPU, CUDA, Vulkan, or Metal
- Download optimized binaries — Only what you need (~100-200 MB)
This keeps the installer tiny while ensuring you get GPU acceleration when available.
- Node.js 20.19+ or 22.12+
- Rust (latest stable)
- System dependencies (see below)
# Install Microsoft Visual Studio C++ Build Tools
# Download from: https://visualstudio.microsoft.com/visual-cpp-build-tools/xcode-select --installsudo apt update
sudo apt install -y libwebkit2gtk-4.1-dev libappindicator3-dev librsvg2-dev patchelf libgtk-3-dev libayatana-appindicator3-dev- Clone the repository:
git clone https://github.com/Icecubesaad/OpenLLM-Studio.git
cd OpenLLM-Studio- Install dependencies:
npm install- Run in development mode:
npm run tauri dev- Build for production:
npm run tauri buildThe built application will be in src-tauri/target/release/bundle/.
OpenLLM Studio uses llama.cpp binaries for local inference. These are downloaded on-demand based on your hardware.
| Variant | Platform | Description |
|---|---|---|
| CPU | All | Pure CPU inference |
| CUDA 12.4 | Windows, Linux | NVIDIA GPUs |
| CUDA 13.1 | Windows, Linux | NVIDIA GPUs (newer) |
| Vulkan | Windows, Linux | AMD/Intel GPUs |
| Metal | macOS | Apple Silicon |
If you prefer to manually download binaries, get them from the official llama.cpp releases page.
- Windows:
%APPDATA%\com.openllm.studio\llama-binaries\ - macOS:
~/Library/Application Support/com.openllm.studio/llama-binaries/ - Linux:
~/.config/com.openllm.studio/llama-binaries/
For accessing private models or increased rate limits:
- Go to Settings
- Navigate to HuggingFace section
- Enter your HF token
- Frontend: React + TypeScript + Tailwind CSS
- Backend: Rust + Tauri v2
- Database: SQLite
- AI Runtime: llama.cpp (on-demand download)
- Model Recommendations: OpenRouter API
OpenLLM-Studio/
├── src/ # React frontend
│ ├── components/ # UI components
│ ├── hooks/ # Custom React hooks
│ ├── pages/ # Page components
│ ├── lib/ # Utilities and types
│ └── store/ # State management
├── src-tauri/ # Rust backend
│ ├── src/ # Rust source code
│ ├── migrations/ # Database migrations
│ └── icons/ # App icons
├── .github/workflows/ # CI/CD pipelines
└── public/ # Static assets
npm run devcd src-tauri
cargo check
cargo testnpm run build- Ensure you have the latest GPU drivers installed
- For NVIDIA: Install CUDA Toolkit
- For AMD: Install latest drivers from AMD Support
- Check Settings → General → Hardware Info
- Check your internet connection
- Try a different variant (Settings → General → Runtime Engine)
- Use CPU variant as fallback
- Check firewall/proxy settings
- Ensure Node.js version is 20.19+ or 22.12+
- Clear node_modules:
rm -rf node_modules && npm install - Clear Rust build cache:
cd src-tauri && cargo clean - Install system dependencies (see above)
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
MIT License - see LICENSE for details.
- llama.cpp - High-performance LLM inference
- Tauri - Desktop application framework
- React - UI framework
- Tailwind CSS - Styling
- HuggingFace - Model hosting