taOS

taOS is a self-hosted AI agent platform for people who want to run assistants on their own computers and keep their conversations, files, and memory there. It brings agent management and local model access into a browser-based desktop.

• Memory across frameworks: taOSmd stores searchable personal and agent memory locally, while agents can switch between frameworks such as SmolAgents, LangChain, and OpenClaw without losing their history, files, or channel connections.
• Shared agent chat: Agents built with different frameworks can participate in the same conversation, with connections to Telegram, Discord, and Slack.
• Compute across devices: Combine PCs, Macs, Raspberry Pi or Orange Pi boards, and Android phones to handle inference and embeddings, with hardware-aware scheduling and cluster monitoring.
• Web desktop and apps: Manage models, agents, documents, files, and plugins in one interface, with an app catalog and layouts for phones and tablets.

The controller runs on Linux and macOS, with Windows supported through WSL2. Workers also run on Windows and Android. Hardware support includes Apple Silicon through MLX, NVIDIA and AMD GPUs, and Rockchip NPUs; CPU workers can handle embeddings and small models.

The project is open source under AGPL-3.0 and can operate fully offline. Cloud models from providers such as OpenAI and Anthropic are optional; choosing them sends requests to those providers. taOSgo is a separate paid remote-access service. Some catalog apps, frameworks, and models haven't been tested on real hardware and may fail to install.