taOS
Self-hosted AI agent OS. Your memory, chat, agents, and files stay on hardware you own, offline by default, cloud by choice. Offline AI memory (taOSmd), self-ho
What is taOS?
Self-hosted AI agent OS. Your memory, chat, agents, and files stay on hardware you own, offline by default, cloud by choice. Offline AI memory (taOSmd), self-hosted multi-framework group chat, a full web desktop + app store, and auto-clustering across the consumer hardware you already have (Orange/Raspberry Pi, Mac mini, gaming PC).
Overview of taOS
What it is
taOS is a self-hosted AI agent operating system that runs on user-owned hardware. It keeps memory, chat, agents, and files offline by default, with cloud as an optional choice. The software is in beta as of 2026-06-02 and is designed for testers running it on their own hardware. It supports a variety of hardware including Orange Pi, Raspberry Pi, Mac mini, gaming PCs, and more.
Key features
- Framework-agnostic agent lifecycle: agents can switch between frameworks (e.g., SmolAgents, LangChain, OpenClaw) without losing memory, files, or configurations.
- taOSmd memory system: a self-hosted knowledge graph that achieves 97.0% end-to-end Judge accuracy on LongMemEval-S, running on a £170 Orange Pi 5 Plus with no cloud dependencies.
- Full web desktop environment with 43 bundled apps, 109 catalog apps, 47 MCP plugins, 17 agent frameworks, and a local model catalog of 120 manifests (LLMs, vision, embeddings, audio, image generation) plus 167k+ searchable models from HuggingFace.
- Supports multiple hardware backends: Apple Silicon (MLX), NVIDIA, AMD, Rockchip NPU, Raspberry Pi, Android phones, and more.
- Offline-first: data stays on user hardware; cloud models are opt-in.
- Open source under AGPL-3.0 with a self-hostable binary mirror and air-gapped install path.
Use cases
- Running AI agents with persistent memory on personal hardware, fully offline.
- Building multi-framework group chat where agents from different frameworks communicate in a single shared channel.
- Deploying a distributed AI compute cluster using spare hardware (old laptops, SBCs, gaming PCs).
- Managing agent memory, files, communication channels, model access, and configuration from a single web dashboard.
Getting started
Controller (server): bash # Debian / Ubuntu / Fedora / Arch / Alpine / macOS, one-line install curl -fsSL https://raw.githubusercontent.com/jaylfc/taOS/master/scripts/install-server.sh | sudo sh For Alpine/postmarketOS, Debian/Ubuntu, Fedora, and Arch, install dependencies first (see README for specific commands). Manual / development: bash pip install -e . python -m uvicorn tinyagentos.app:create_app --factory --host 0.0.0.0 --port 6969 Open http://your-host:6969 (or http://taos.local:6969 with mDNS).
License and pricing
The license is AGPL-3.0. The software is open source.
Based on the project's public README and repository metadata.