Summary DimensionalOS (dimos) is an agent-native operating system for generalist robotics, aiming to standardize physical space application development. It enables Python applications on diverse hardware (humanoids, quadrupeds, drones) without ROS, facilitating natural language control ("vibecode") and multi-agent systems. dimos integrates robot hardware as native modules, processing streams from perception to control loops, and offers advanced features like SLAM, autonomous exploration, agentive control (MCP), and spatio-temporal memory (RAG).

How It Works dimos employs an agent-native architecture where agents act as core modules subscribing to embedded data streams. It uses Python for application development and abstracts hardware complexities. Communication relies on LCM, with support for SHM, DDS, and ROS 2 transports. "Blueprints" define module composition and wiring via autoconnect, enabling dynamic remapping and transport overrides for modularity.

Quick Start & Requirements Installation: curl -fsSL https://raw.githubusercontent.com/dimensionalOS/dimos/main/scripts/install.sh | bash. Requires Python 3.12. A typical setup uses uv:

uv venv --python "3.12"
source .venv/bin/activate
uv pip install 'dimos[base,unitree]'

Supports NixOS, Linux, macOS; CUDA for GPU acceleration; Docker. Hardware integrations are tiered (🟩 stable, 🟨 beta, 🟧 alpha, 🟥 experimental), including Unitree, Xarm, AgileX, DJI, MAVLink. Full system requirements: docs/requirements.md.

Highlighted Details

• Agent-native design with natural language robot control ("vibecode").
• Unified Python SDK for humanoid, quadruped, and drone robotics.
• Local and hosted multi-agent system support.
• Advanced perception, agentive control (MCP), and spatio-temporal memory (RAG).
• Integrated navigation and mapping (SLAM, autonomous exploration).
• Extensible communication supporting LCM, SHM, DDS, and ROS 2.

Maintenance & Community Community engagement via Discord (https://discord.gg/dimos). GitHub badges indicate repository activity. Core maintainers, sponsorships, or a public roadmap are not detailed in the README excerpt.

Licensing & Compatibility The provided README content does not specify the software license. This omission requires further investigation for usage restrictions, especially for commercial applications.

Limitations & Caveats The project is explicitly labeled "Pre-Release Beta," indicating potential instability. Hardware support is tiered by stability, with many integrations not yet stable. The absence of a stated license is a significant adoption blocker.