Browse and download AI models to run directly on your device. No internet required after download.
Full precision 1B model from Meta
1B model with SpinQuant quantization (recommended)
Full precision 3B model from Meta
3B model with SpinQuant quantization
Lightweight Qwen 3 model with thinking capability
Larger Qwen 3 model with thinking capability
Large Qwen 3 model with thinking capability
Ultra-lightweight Qwen 2.5 model
Lightweight Qwen 2.5 model
Mid-size Qwen 2.5 model
Ultra-lightweight model optimized for function calling
Lightweight model optimized for function calling
Mid-size model optimized for function calling
Ultra-lightweight model for simple tasks
Small but capable model
Larger SmolLM model with better capabilities
Microsoft's Phi 4 Mini model
Lightweight text embedding model for semantic search and RAG
Lightweight Qwen 3 model for llama.cpp with thinking capability
Text embedding model for RAG using llama.cpp
Download these models directly in the app. Available on Android and Web.
Open App to Download