# stepjam/RLBench

A large-scale benchmark and learning environment.

Repository: https://github.com/stepjam/RLBench
Canonical: https://ross.abutalabs.com/products/rlbench
Homepage: https://sites.google.com/corp/view/rlbench
Language: Python
License: NOASSERTION
License Family: other
Last push: 2025-01-25T13:13:19+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 3, release rhythm 8, longevity 100
- inputs: {"age_days": 2533, "days_push": 585, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- 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 1811, forks 321 (observed 2026-08-28T04:05:39.530379+00:00)

## What it is
RLBench is a large-scale robot learning benchmark and simulation environment built on CoppeliaSim and PyRep, featuring 100 hand-designed manipulation tasks. It supports research in vision-guided manipulation including reinforcement learning, imitation learning, multi-task learning, and few-shot learning.

## Use cases
- benchmark robot manipulation policies in simulation
- train reinforcement learning agents on robotic tasks
- collect demonstrations for imitation learning
- run multi-task and few-shot meta-learning experiments
- evaluate vision-guided grasping and manipulation
- build custom robot manipulation tasks

## When to choose
- you need a diverse set of simulated manipulation tasks with ground-truth data
- you research RL, imitation learning, or few-shot learning for robotics
- you want a Gym-compatible robot learning environment

## When to avoid
- you need real-robot hardware experiments only
- you work outside Linux or cannot install CoppeliaSim
- you need lightweight or non-robotics simulation

## Facets
- artifact type: dataset
- maturity: active
- function: simulation, machine-learning, reinforcement-learning, computer-vision, benchmarking
- domain: robotics, reinforcement-learning, machine-learning, computer-vision, simulation
- platform: python
- tags: robot-learning, manipulation, imitation-learning, benchmark, coppeliasim, few-shot-learning, gym-environment, linux

## Member repositories
- stepjam/RLBench (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:39.530379+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 2026-08-30T03:21:16.853862+00:00, confidence not recorded.
  - readme: https://github.com/stepjam/RLBench (fetched 2026-08-28T04:05:39.530379+00:00, sha 77ea470aaaf8)
  - homepage: https://sites.google.com/corp/view/rlbench (fetched 2026-08-29T11:00:11.335712+00:00, sha d2810d1d83d5)
- Data as of 2026-08-30T08:39:29.467469+00:00.