Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.lmsa.app/llms.txt

Use this file to discover all available pages before exploring further.

LMSA (Local Model Smart Assistant) is a premium Android app that gives you a full-featured AI chat interface — without subscriptions, without sending your data to third-party servers, and without sacrificing capability. You connect it to an AI provider of your choice: a local server running on your own PC, or a cloud API you control, and LMSA handles the rest.

How LMSA connects to AI

LMSA supports three types of AI providers. You choose one in the app’s settings.

LM Studio

Connect to a GGUF model running locally on your PC or Mac via LM Studio. Your conversations never leave your home network.

Ollama

Connect to any model served by an Ollama instance on your local network. Supports all Ollama-compatible models.

OpenRouter

Access 100+ leading cloud models — including Llama, Mistral, DeepSeek, Claude, and Gemini — using your own OpenRouter API key.

Custom endpoint

Point LMSA at any OpenAI-compatible API endpoint with an optional API key. Works with self-hosted or enterprise deployments.

Key benefits

Privacy first. When you use a local provider, your conversations stay entirely on your device and local network. LMSA collects no telemetry and tracks no conversations. When using OpenRouter, your requests go directly to the provider — there are no middleman servers. No subscriptions. Download LMSA for free and use the core features immediately. Upgrade once with a one-time in-app purchase to remove ads and unlock all premium capabilities — including Character Card generation and unlimited image generation — forever. Any model, any time. Switch between a locally loaded GGUF model on your PC and 100+ cloud models without leaving the app. LMSA makes model-switching instant. Built for mobile. LMSA renders LaTeX math, highlights code syntax, plays AI responses aloud with Text-to-Speech, supports biometric lock, and scales across phones and tablets.

System requirements

RequirementDetails
Android version6.0 (Marshmallow) or higher
Local server modeYour phone and PC must be on the same Wi-Fi network
Cloud modeActive internet connection; OpenRouter API key required
StorageVaries by usage — conversations are stored on-device
LMSA does not run AI models on your phone. It acts as a lightweight interface that connects to a model running on your PC (local mode) or a cloud API (cloud mode). Your phone’s processing power does not affect model performance.

What you can do with LMSA

  • Chat with any LLM using custom system prompts and persona templates, including full Character Card V2 support
  • Attach images, PDFs, and documents for context-aware analysis
  • Generate images directly in chat with the /image command
  • Search the web in real time with Brave Search integration
  • Listen to AI responses with native Text-to-Speech
  • Secure your chat history with biometric (fingerprint or face) unlock
  • Save and switch between multiple server connection presets
  • Export chat histories for backup or analysis

Ready to install?

Install LMSA

Download from Google Play and learn about free vs. premium.

Quick start

Connect to your first AI provider and send your first message.