What is anythingllm-mobile?
AnythingLLM Mobile brings AnythingLLM's local-first philosophy to mobile phones. We believe that intelligence should be private, instant, and usable on every device you own, with all the context you want it to have.
Mobile devices, even the most advanced ones, are limited before running models due to overheating or insufficient memory, therefore AnythingLLM Mobile adopts a hybrid approach: your phone is the carrier of the agent, while the model can exist in the place that is most suitable for you.
Key features
Small language models (SLM) can run on mobile devices and can be paired via the web with LM Studio, AnythingLLM Desktop, or llama.cpp, or connected to the cloud provider you are already using.
Supports RAG (Retrieval-Augmented Generation) operations on the device, including documents in PDF, Word, Excel and other formats, with embedding, vector database and re-ranking all completed on the device.