# Genesis-Embodied-AI/RoboGen

A generative and self-guided robotic agent that endlessly propose and master new skills.

Repository: https://github.com/Genesis-Embodied-AI/RoboGen
Canonical: https://ross.abutalabs.com/products/robogen
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-05-31T01:32:29+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 74
- inputs: {"age_days": 1037, "days_push": 825, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1223, forks 108 (observed 2026-08-28T04:04:02.569035+00:00)

## What it is
RoboGen is a self-guided generative robotic agent that autonomously proposes new tasks, generates simulation environments, and learns robotic skills via reinforcement learning. It is a Python research framework built on PyBullet (with a planned Genesis simulation engine backend), published at ICML 2024.

## Use cases
- automatically generate robot manipulation and locomotion tasks
- train robotic skills with reinforcement learning in simulation
- use LLMs to propose and decompose robotics tasks
- generate simulation environments for robot learning
- research automated curriculum generation for robots
- run one-click task generation and skill learning pipelines
- benchmark generative simulation approaches for robotics

## When to choose
- you are researching automated task generation or lifelong robot skill learning
- you want an LLM-driven pipeline that proposes tasks, builds environments, and trains policies
- you need a PyBullet-based reimplementation of the RoboGen ICML 2024 paper
- you want to experiment with generative simulation for rigid-body manipulation and locomotion

## When to avoid
- you need a production-ready robotics platform or stable API
- you require soft-body manipulation or the full Genesis-powered pipeline, which is not yet released
- you need guaranteed maintenance or commercial support
- you want a lightweight tool without GPU, simulation, and LLM dependencies

## Facets
- artifact type: library
- maturity: experimental
- function: machine-learning, simulation, agent-framework, llm-inference, reinforcement-learning
- domain: robotics, artificial-intelligence, simulation, machine-learning
- platform: python
- tags: generative-simulation, robot-learning, llm-agent, pybullet, task-generation, skill-acquisition, research-code, icml-2024, linux, gpu

## Member repositories
- Genesis-Embodied-AI/RoboGen (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:02.569035+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 2026-08-30T06:15:30.887480+00:00, confidence not recorded.
  - readme: https://github.com/Genesis-Embodied-AI/RoboGen (fetched 2026-08-28T04:04:02.569035+00:00, sha 53888c7ae55e)
- Data as of 2026-08-30T08:39:29.467469+00:00.