#13 · Primary category: Robotics & Embodied AI Simulation

dimos

Dimensional is the agentic operating system for physical space. Command humanoids, quadrupeds, drones, and other hardware platforms in natural language and build multi-agent systems that work seamlessly with physical input (cameras, lidar, actuators).

Project last updated:08/29/26

GitHub Stars

4.4K

Forks

796

Contributors

43

License

Other

Why we included this project

Robotics teams and embodied-AI developers who want to drive real hardware with natural language have a practical starting point here. Dimensional ships a full software stack for humanoids, quadrupeds, arms, and drones that covers SLAM navigation, perception, spatial memory, and agentic control, with no ROS required. You can run the same blueprints against MuJoCo simulations or replayed sessions when no robot is on the desk, which makes prototyping agent behaviors and MCP tool calls much less risky before anything physical is involved. The CLI and MCP server let you plug existing agent tools into robot skills, and the documented Unitree, XArm, and drone integrations give concrete targets to aim at. It is pre-release beta, so expect rough edges, but the breadth of working blueprints makes it a useful reference when evaluating an agent-native robotics SDK.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category