AnythingLLM Mobile is currently in closed beta.
Introduction
this documentation is currently in progress while we work on a public release - it may be incomplete or incorrect
AnythingLLM Mobile is a mobile app that brings the entire AnythingLLM experience onto your phone.
It is currently in closed beta for Android. You can join the beta by filling out this form (opens in a new tab) and joining the #anythingllm-mobile channel in the AnythingLLM Discord (opens in a new tab).
Features
- Chat with local SLM - Chat with your local SLM (small language model) on your phone. Supports both reasoning and non-reasoning models.
- Change models on the fly - Easily swap between different models
- Workspace and Threads - Create workspaces and threads to organize your chats
- On device RAG - Locally process your documents and use them in your chats all fully offline
- Agentic Tools - Leverage the power of AnythingLLM's agentic tools like web search, web scraping, deep research, and even cross app interactions like drafting emails or managing your calendar
- Sync with AnythingLLM Desktop & Cloud - Sync your chats, workspaces, and threads with AnythingLLM Desktop or AnythingLLM Cloud/Self-hosted instances
For Beta Testers
If you are a beta tester, you should have received an email with a link to the app to download from Google Play or via Direct Download (APK).
DO NOT share the app with anyone outside of the beta testers. This is a closed beta and external access will only slow down the development process.
If you have any general questions, please join the #anythingllm-mobile channel in the AnythingLLM Discord (opens in a new tab) and we'll help you out.
Feedback Reporting
All feedback should be officially reported via the AnythingLLM Feedback Form (opens in a new tab).
Public Issue Tracking
All public issues should be reported via the AnythingLLM Mobile Beta Issue Tracker (opens in a new tab).
Common Questions
IOS support?
We are planning to support iOS in the future. Currently, we are focusing on Android for a full release by the end of September 2025. iOS support coming after that in October 2025.
Can I download any model I want?
Right now, for performance reasons, we only support a hand-picked models. Eventually we will support any model you want, but for now, we are focusing on performance and stability.
How does syncing with AnythingLLM Desktop & Cloud work?
Requires version 1.8.5 or higher of AnythingLLM Desktop or AnythingLLM Cloud. In 1.8.5, this feature is hidden behind the "Experimental features" sidebar item.
AnythingLLM Mobile while functional and complete standalone, is designed to be also be a companion to AnythingLLM Desktop and AnythingLLM Cloud.
Since mobile devices have limited resources, we now have the ability to sync your chats, workspaces, and threads with AnythingLLM Desktop or AnythingLLM Cloud/Self-hosted instances in addition to being able to delegate inference across your local network or cloud instances for more compute and more powerful models, but in the mobile form factor!
This technology is called Distributed InferenceTM and is a key part of AnythingLLM's vision for the future of local AI.
How does the on device RAG work?
AnythingLLM Mobile runs a small embedding model + local vector database on your device to provide RAG capabilities with citations.
How can I add my own agent tools?
Currently, to use custom agent tools, MCPs or otherwise, you should use the sync feature with AnythingLLM Desktop or AnythingLLM Cloud. Customization of agent tools on mobile standalone is not yet supported.