---
title: RLBench
description: "100 hand-designed vision-guided tasks on CoppeliaSim / PyRep. WorldFoundry imports official rollouts — it does not run the simulator."
---

RLBench (IEEE RA-L / ICRA 2020) is Imperial College London's Dyson Robotics Lab benchmark: **100** unique, hand-designed manipulation tasks for a **Franka Panda** in **CoppeliaSim** (via **PyRep**). Each task has procedural variations (object poses, goals). A motion planner can emit **unlimited** demonstrations, so the same suite supports imitation learning, RL, and a **few-shot** protocol. Observations are multi-camera vision plus state / language.

Paper: [RLBench: The Robot Learning Benchmark & Learning Environment](https://arxiv.org/abs/1909.12271). Project: [sites.google.com/view/rlbench](https://sites.google.com/view/rlbench). Code: [stepjam/RLBench](https://github.com/stepjam/RLBench). GitHub license metadata in this catalog is `other:noassertion_github_api` — read the upstream terms before reuse.

WorldFoundry is **official-results only**. You run CoppeliaSim / PyRep / the official task assets upstream, then `official-validation` normalizes JSON/CSV. No in-tree simulator. Declare the task list and the protocol (few-shot vs IL vs RL) with the export.

<BenchmarkPageSections id="rlbench" locale="en" />

## How to run

```bash
worldfoundry-eval zoo benchmark-run \
  --benchmark-id rlbench \
  --mode official-validation \
  --official-results-path <official_results.json> \
  --output-dir tmp/rlbench/official-validation \
  --json
```

Default import path when you omit `--official-results-path` is `WORLDFOUNDRY_RLBENCH_RESULTS_PATH`. Manifest: `worldfoundry/data/benchmarks/tasks/external/rlbench.yaml`. Outputs: `scorecard.json` (primary `success_rate`) and `raw_results.jsonl`.

[← Benchmark Hub](/docs/evaluation/benchmark-hub)