# AgibotTech/genie_sim

Simulation Platform from AgiBot

Repository: https://github.com/AgibotTech/genie_sim
Canonical: https://ross.abutalabs.com/products/genie_sim
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-08-18T06:51:29+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 64, longevity 35
- inputs: {"age_days": 499, "days_push": 15, "days_rel": 29, "gap_med": null, "n_releases_24m": 1}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1359, forks 116 (observed 2026-08-28T04:04:29.733685+00:00)

## What it is
Genie Sim is AgiBot's open-source simulation platform for embodied intelligence, providing environment reconstruction, LLM-driven scene generation, synthetic data collection, and standardized benchmarking of robot policies. It ships with a `geniesim` CLI, ROS 2-based realtime engine, and agent-ready SKILL recipes for driving benchmarks and teleoperation.

## Use cases
- simulate humanoid robot tasks for embodied AI research
- benchmark robot manipulation policies across hundreds of tasks
- generate synthetic training data for robot learning
- reconstruct real-world 3D scenes for simulation
- teleoperate a simulated robot via ROS 2
- evaluate vision-language-action models without physical hardware

## When to choose
- you need high-fidelity robot simulation with ROS 2 integration
- you want standardized evaluation of embodied AI models
- you need large-scale synthetic data for robot policy training
- you work with AgiBot robots or Isaac Sim-based pipelines

## When to avoid
- you need lightweight 2D physics simulation
- you only need a game engine for entertainment purposes
- you require a mature commercial simulator with vendor support
- your workflow does not involve robotics or embodied AI

## Facets
- artifact type: framework
- maturity: active
- function: simulation, robotics, machine-learning, benchmarking, data-generation, llm-inference
- domain: robotics, simulation, artificial-intelligence, machine-learning, autonomous-vehicles
- platform: python
- tags: embodied-intelligence, isaac-sim, ros2, robot-simulation, synthetic-data, teleoperation, benchmark, scene-generation, 3d-reconstruction, linux, docker, ros

## Member repositories
- AgibotTech/genie_sim (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:29.733685+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 2026-08-30T04:41:40.803016+00:00, confidence not recorded.
  - readme: https://github.com/AgibotTech/genie_sim (fetched 2026-08-28T04:04:29.733685+00:00, sha 4956ad65533d)
- Data as of 2026-08-30T08:39:29.467469+00:00.