# ARISE-Initiative/robomimic

robomimic: A Modular Framework for Robot Learning from Demonstration

Repository: https://github.com/ARISE-Initiative/robomimic
Canonical: https://ross.abutalabs.com/products/robomimic
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-09T22:06:47+00:00

## Health v2 (maintenance only)
Score: 69/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 96, release rhythm 16, longevity 100
- inputs: {"age_days": 1853, "days_push": 24, "days_rel": 433, "gap_med": 107, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1535, forks 416 (observed 2026-08-28T04:04:59.709921+00:00)

## What it is
robomimic is a modular Python framework for robot learning from demonstration, providing standardized demonstration datasets and offline learning algorithms for robot manipulation. It enables researchers to fairly benchmark tasks and algorithms in a reproducible way.

## Use cases
- train robot manipulation policies from human demonstrations
- benchmark imitation learning algorithms on standardized datasets
- run offline reinforcement learning experiments like IQL
- train diffusion policy models for robot control
- evaluate language-conditioned robot policies
- reproduce robot learning research results

## When to choose
- you need a reproducible benchmark suite for robot learning from demonstration
- you want ready-made manipulation datasets and offline RL algorithms
- you work with robosuite or MuJoCo simulation environments

## When to avoid
- you need real-time online RL training with live robot interaction
- you want a general-purpose deep learning framework unrelated to robotics
- you need plug-and-play deployment on physical robots without research setup

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, deep-learning, simulation, benchmarking
- domain: robotics, machine-learning, artificial-intelligence
- platform: python
- tags: robot-learning, imitation-learning, offline-rl, learning-from-demonstration, manipulation, robosuite, diffusion-policy, research, linux, macos

## Member repositories
- ARISE-Initiative/robomimic (main) score 69

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:59.709921+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 2026-08-30T04:31:07.226534+00:00, confidence not recorded.
  - readme: https://github.com/ARISE-Initiative/robomimic (fetched 2026-08-28T04:04:59.709921+00:00, sha 556dfe48d2a8)
  - registry_pypi: https://pypi.org/pypi/robomimic/json (fetched 2026-08-29T11:32:59.792693+00:00, sha 16ea92fba532)
- Data as of 2026-08-30T08:39:29.467469+00:00.