LMSA is a premium, feature-rich Android interface designed for seamless interaction with Large Language Models (LLMs). If you're running local models via LM Studio and Ollama or leveraging cloud power through OpenRouter, LMSA provides a state-of-the-art chat experience with a focus on privacy, performance, and rich feature sets.
- Multi-Provider Support: Seamlessly connect to LM Studio, Ollama, OpenRouter, or any OpenAI-compatible custom endpoint.
- Model Context Protocol (MCP): Native support for LM Studio MCP servers and
mcp.jsonplugins for extended model capabilities. - Real-Time Web Search: Augment AI responses with live data using the Brave Search API (with SearXNG fallback).
- Multimodal & Vision: Full support for multimodal models—upload images for analysis or use the built-in OCR (via Tesseract.js) to extract text from documents.
- AI Image Generation: Create images directly in your chats using the
/imagecommand. - Advanced Reasoning Controls: Toggle "Thinking" visibility and adjust "Reasoning Level" (Thinking Effort) for supported models.
- Persona & Character Templates:
- Library of Personas: Start chats with specialized contexts like Math Tutor, Code Assistant, and more.
- Custom Templates: Create and save your own system prompts with custom names and avatars.
- Character Card V2 Support: Full support for the industry-standard Character Card V2 spec, enabling complex roleplay with personality, scenario, and dialogue examples.
- AI-Powered Generation: Automatically generate high-quality system prompts or complete character cards using AI.
- Smart Conversational Tools:
- Smart Reply (Beta): Get AI-powered quick reply suggestions.
- Auto-Titles: Let the AI automatically name your chats based on context.
- Privacy & Security:
- Biometric Unlock: Secure your chats with device-level fingerprint or PIN authentication.
- Offline Access: Full support for LAN-only mode when using local servers.
- Rich UI/UX:
- LaTeX & Code Highlighting: Beautifully rendered math (KaTeX) and syntax-highlighted code blocks.
- Customization: Change chat fonts, sizes, and toggle model name displays.
- Native TTS: High-quality Text-to-Speech playback using Android's native engine.
- Power User Tools:
- Connection Presets: Save and switch between multiple server configurations effortlessly.
- Deep Integration: Import LM Studio profiles, Character Card V2 specs, and manage custom persona templates.
- Data Portability: Export full chat histories and character data.
- Android Native: Built with Kotlin and modern Android Jetpack components.
- Hybrid Core: High-performance WebView implementation for a rich, responsive UI.
- Modern Web UI: Styled with a custom, premium design system inspired by Tailwind CSS.
- Open Source Libraries:
- KaTeX (Math rendering)
- Highlight.js (Syntax highlighting)
- Tesseract.js (OCR)
- PDF.js (PDF viewing)
LMSA is available for download on the Google Play Store:
- Android Studio Ladybug or later.
- Android SDK 34+.
- A local LLM server (optional, for local mode).
- Clone the repository:
git clone https://github.com/TechMitten/LMSA.git
- Open the project in Android Studio.
- Sync Project with Gradle Files.
- Build and run on your device or emulator.
LMSA makes it easy to migrate your workflows:
- LM Studio Profiles: Import your system prompts and configurations directly.
- Character Cards: Full support for
.jsonand image-embedded Character Card V2. - Persona Templates: Create, edit, and manage your own reusable chat personas with custom avatars.
- Chat History: Export your conversations for backup or analysis.
This project is licensed under the PolyForm Noncommercial License 1.0.0.
Caution
Commercial Use Prohibited You are free to clone, modify, and use this software for personal or educational purposes. However, any commercial use, including selling this software or its derivatives, is strictly prohibited. See the LICENSE file for the full legal text.
Built with ❤️ by the LMSA Team.
(Google Play and the Google Play logo are trademarks of Google LLC.)