Avatar is an AI harness for running your agents in controlled environments. Avatars are agents that primarily interact with real users and the physical world. It comes with an inbuild memory module, that with persistant storage.
The local memory can be used either as episodal or transactional.
Use local ai in your desktop or in your private cloud, with no worries about privacy or data leakage.
The Ai Chat page provides a UI, with rich features for interacting with LLM’s on your local machine.
Avatar supports Rest API, GRPC and also comes with an inbuild ui.
The UI can be access by default are port 9080.
http://localhost:9080
Local Models
For local ai inference use any of the following
- llama.cpp
- ollama
- mlx serve (only on macos)
Since Avatar supports OpenAIAPI, any of the cloud AI vendors like Calude, ChatGTP can be used, by configuring the endpoints, and the token.