# OpenMind/OM1

Modular AI HAL (Hardware Abstraction Layer) for Robots

Repository: https://github.com/OpenMind/OM1
Canonical: https://ross.abutalabs.com/products/om1
Homepage: https://openmind.com
Language: Go
License: MIT
License Family: permissive
Topics: llm, multiagent, robotics, ros2, zenoh
Last push: 2026-08-25T22:19:18+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 81, longevity 43
- inputs: {"age_days": 602, "days_push": 8, "days_rel": 127, "gap_med": 20.5, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2897, forks 991 (observed 2026-08-28T04:07:28.756892+00:00)

## What it is
OM1 is a modular AI runtime and hardware abstraction layer for building multimodal AI agents that run on physical robots and in simulators. Written in Go for performance, it connects sensors like cameras and LIDAR to LLM/VLM-powered reasoning and physical actions such as navigation and speech.

## Use cases
- build multimodal AI agents for humanoid or quadruped robots
- connect LLMs to robot hardware via ROS2 or Zenoh
- add voice conversation and vision to a robot
- run autonomous navigation with camera and LIDAR input
- simulate robot agents in Gazebo or Isaac Sim
- monitor AI pipeline latencies with Prometheus and Grafana

## When to choose
- you want a modular, plugin-based runtime to connect LLMs to robot hardware
- you need low-latency, edge-friendly performance in a single Go binary
- you target multiple robot form factors or simulators and want easy reconfiguration
- you want pre-configured endpoints for many LLM/VLM/TTS providers including local Ollama

## When to avoid
- you need the full feature set of the deprecated Python runtime, parts of which are still missing in Go
- your project has nothing to do with robotics or embodied agents
- you require a mature, long-stable production framework rather than a fast-moving one

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, speech-recognition, tts, middleware, plugin-system, monitoring
- domain: robotics, artificial-intelligence, large-language-models, autonomous-vehicles
- platform: go, python
- tags: robotics-hal, hardware-abstraction-layer, multimodal-agents, zenoh, ros2, humanoid-robots, simulators, edge-devices, ai-agents, linux, docker, gpu

## Member repositories
- OpenMind/OM1 (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:28.756892+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:35:13.567454+00:00, confidence not recorded.
  - readme: https://github.com/OpenMind/OM1 (fetched 2026-08-28T04:07:28.756892+00:00, sha f6a3d43a411d)
  - homepage: https://openmind.com (fetched 2026-08-29T09:50:00.386980+00:00, sha 1e924efb62dc)
- Data as of 2026-08-30T08:39:29.467469+00:00.