🚀 Pocket LLM v1.4.0
Pocket LLM now ships as a much smaller base APK and lets users download models directly inside the app after installation.
✨ Highlights
- 📦 Smaller base APK of around 200 MB
- ⬇️ Models are no longer bundled inside the APK
- 📱 First-launch model picker with on-device downloads
- 📚 Support for multiple downloaded models
- 🔁 In-app model switching
- 🧠 Collapsible retained thinking text for supported models
- 🎨 Basic UI improvements and polish
🤖 Supported models
- 💎 Gemma 4 E4B LiteRT - Best for flagship mobiles
- ⚖️ Gemma 4 E2B LiteRT - Best for decent to mid-range mobiles
- 📱 Qwen3 0.6B LiteRT - Best for low-end mobiles
- ⚡ Qwen3 0.6B Q4F16 ONNX - Good for low to mid-range mobiles
- 🧠 Qwen2.5 0.5B ONNX - Best for mid to high-end mobiles, full precision
📝 Notes
- 🌐 Internet is needed only to download models
- 🔒 Once a model is installed, chat stays fully on-device