About AlphaAvatar
AlphaAvatar is an open source, self hostable runtime for a real time personal AI assistant that can talk, see, remember, plan and act. It focuses on the personal assistant layer: real time voice and vision interaction, persistent user context, and an optional real time virtual avatar, connected to tools and long running workflows across a user's workspace and daily life.
The project is aimed at developers and researchers building personal assistants and multimodal agents. It is organized around composable agent plugins for memory, persona, reflection, planning, external tools, RAG, deep research, messaging channels and real time virtual character interaction, with MCP support for tools. In practice the assistant can manage personal knowledge, retrieve notes, track life metrics, schedule tasks, call tools, research information and keep continuity across conversations and modalities. The site states plainly that it is not a replacement for general chat models, coding agents or desktop automation tools. Version 0.6.0 added a status plugin, sampled visual inputs and status aware tool feedback. A live demo, documentation, a Discord community and GitHub discussions are available.
AlphaAvatar is free and released under the Apache 2.0 license. No paid plan, hosted tier or pricing page is published.
The site's metadata still describes it as an avatar creation and animation platform, but the current product is an assistant runtime in which the avatar is one component. Despite its listing among coding tools, it is not a coding assistant.