Private AI chat on your Android phone.
QuantLM brings fast, local AI experiences to mobile with a clean chat interface, vision support, and flexible model options. It is designed for everyday use, with privacy and control built in.
- Chat with AI directly on your device
- Ask questions about photos and screenshots
- Use voice input for faster conversations
- Choose the model that matches your device and needs
- Keep your chats and settings on your phone
- Natural multi-turn conversations
- Streaming replies for a responsive feel
- Conversation history so you can continue where you left off
- Attach images from camera or gallery
- Ask questions about what is in an image
- Use quick actions like Describe, Read Text, Analyze, Identify, and Translate
- Speak your prompt with microphone input
- Listen to responses with text-to-speech
- Browse and download models in-app
- Track progress with clear download status
- Import local model files
- Switch models anytime from the app
- Optional app lock (PIN, password, pattern, biometric)
- Theme options (system, light, dark)
- No forced account setup for normal use
QuantLM includes a curated model library from popular families such as:
- Phi
- Qwen
- Llama
- SmolLM / SmolVLM
- DeepSeek
This gives you options for speed, quality, and multimodal use depending on your device.
- Open QuantLM.
- Go to Models and download one model.
- Load the model and start chatting.
- Optionally attach an image or use voice input.
QuantLM is designed for on-device use. Your conversations, downloaded models, and settings remain on your phone unless you explicitly choose to share content.
QuantLM only requests permissions needed for features you use, such as camera (image chat), microphone (voice input), notifications (download updates), and biometrics (app lock).
You can install QuantLM from this repository's release APK builds.
Suggestions and bug reports are welcome via GitHub Issues.
Custom Source-Available License. See LICENSE.