| Google Play Store | GitHub Releases |
|---|---|
 |
 |
🎉 Latest Update: Version 5.9 - Now Available!
- Fixed Photo Attachments for Gemma Models: Resolved issue where some users could not attach photos to Gemma models (thank you toiletman74 for the report)
- Removed the autoscroll feature: Some users were complaining that the auto scroll feature made it difficult to read messages while text was being generated.
- Fixed Character Display Issues: Resolved broken character feature bug (thank you teneykio for the report)
- Enhanced Side Menu: Made improvements to the side menu for better visibility of the saved chats list
- Streamlined User Experience: Removed the "What's New" menu feature that would force it to show every time the app loads
- Improved Data Caching: Enhanced the way the app caches data to improve how updates are implemented through the Play Store
- Mobile UI Enhancements: Made minor UI improvements for smaller mobile screens
- Vision Language Model Support: Added full support for vision language models (VLMs). You can now upload and analyze images alongside text, enabling multimodal conversations with AI models that can understand and discuss visual content.
- Fixed PDF File Attachments: Resolved the issue where PDF files weren't allowed as attachments. You can now upload and analyze PDF documents seamlessly with your AI models.
LMSA (LM Studio Assistant) is an Android front-end application for LM Studio that provides a clean, user-friendly interface to interact with language models on your Android device. It's designed with privacy in mind, offering a tracking-free experience for users who want to leverage the power of large language models on mobile.
- Connect to LM Studio running on your computer
- Chat with AI models through a mobile-optimized interface
- Upload and analyze documents using AI
- Upload and analyze images with vision language models
- Customize AI behavior with system prompts and temperature settings
This project has transitioned to open source development.
The project is now community-driven and open source. Development efforts are focused on collaborative community contributions, with regular updates and improvements available through GitHub releases.
- Community updates improving app security and stability
- Privacy-focused design with no unnecessary tracking
- Secure connections to your LM Studio instance
- Privacy First - Your conversations never leave your personal network as models run locally
- Document analysis - Chat with your files
- Image analysis - Upload and discuss images with vision language models
- Multimodal interactions - Combine text, documents, and images in conversations
- Custom instructions - Personalize AI responses
- Precision tuning - Adjust response creativity and accuracy
- System prompt - Set context for more relevant responses
- AI Reasoning Visibility - Toggle the ability to see the model's "thinking" process before generating responses
- File Processing - Upload and process files to include in your prompts for more context-aware responses
- Clean interface - Intuitive design for effortless interaction
- Model information - View details about the loaded AI model
- Tablet optimization - Enhanced layout for larger screens
- Sidebar navigation - Quick access to conversations and settings
- Dark Mode Support - Chat comfortably day or night with full UI theme support
- Mobile-Responsive Design - Optimized for both phones and tablets of various screen sizes
- Quick Navigation - "Scroll to Bottom" button for faster movement in long chats
- Multiple Conversations - Create and manage separate chat threads for different topics
- Comprehensive History - Save, browse, and continue previous conversations with organized chat management
- Automatic Titles - Saved chats are automatically titled for easy identification
- Import/Export - Easily backup and transfer your saved conversations between devices
- Response Management - Copy, regenerate, or modify AI responses with ease
- Simple Connection - Connect to your LM Studio server with just an IP address and port
- Advanced Customization - Adjust temperature, system prompts, and other model parameters directly from your phone
- Start LM Studio on your computer and load your favorite language model (including vision language models)
- Activate the server feature in LM Studio (usually on port 1234)
- Connect the Android app to your computer using your local network
- Start chatting with your AI models from anywhere in your home
LMSA connects to LM Studio running on your computer, allowing you to:
- Access powerful AI language models from your mobile device
- Chat with AI models using a simple, intuitive interface
- Upload documents for the AI to analyze and discuss
- Upload images for vision language models to analyze and discuss
- Customize AI behavior through temperature and system prompt settings
- AI enthusiasts who want to access their models on the go
- Privacy-conscious users who prefer keeping their data local
- Developers testing different prompts and model responses
- Researchers who need to include file content in their AI interactions
- Visual content creators who want to analyze and discuss images with AI
- Anyone who wants the convenience of mobile access to powerful AI
- Android 5.0 or higher
- LM Studio installed and running on a computer with a suitable language model (text or vision)
- Both devices connected to the same network
- Download and install the app
- Start LM Studio and load a model on your computer
- Start the LM Studio server
- Open the app and enter the server IP and port in Settings
- Begin chatting immediately!
LMSA - Google Play
Publisher: IslandApps
Updates: Regular feature, security and maintenance updates
Download Now →
LMSA is a third-party application and is not affiliated with LM Studio or its developers. This app is independently developed to provide an Android front-end interface for interacting with LM Studio. Use of this app is at your own discretion, and the developers of LMSA are not responsible for any issues arising from its use.
© 2025 IslandApps. All rights reserved. Google Play is a trademark of Google LLC.