# VLABench (/docs/evaluation/benchmark-hub/vlabench)



## About [#about]

VLABench external benchmark manifest for language-conditioned manipulation policy rollouts.

## Metrics [#metrics]

VLABench external benchmark manifest for language-conditioned manipulation policy rollouts.

### Task YAML [#task-yaml]

`worldfoundry/data/benchmarks/tasks/external/vlabench.yaml`

### Primary [#primary]

* `success_rate`

| Metric            | Focus                                                                              |
| ----------------- | ---------------------------------------------------------------------------------- |
| `success_rate`    | Share of episodes or tasks that complete successfully.                             |
| `task_success`    | Share of episodes or tasks that complete successfully.                             |
| `episode_success` | Share of episodes or tasks that complete successfully.                             |
| `reward`          | Cumulative reward on VLABench language-conditioned tabletop manipulation rollouts. |

## Environment setup [#environment-setup]

### Runtime environment [#runtime-environment]

* **Conda env:** `vlabench-runtime`.
* **Separate setup:** this benchmark expects a dedicated conda env rather than the unified runtime.
* WorldFoundry provides contract and official result normalization for caller-provided VLABench outputs.

### Evaluation data [#evaluation-data]

* Candidate model outputs: set `WORLDFOUNDRY_VLABENCH_RESULTS_PATH` to the generated-video or rollout artifact root.
* `${WORLDFOUNDRY_DATA_DIR}/datasets/vlabench`
* `${WORLDFOUNDRY_CACHE_DIR}/repos/OpenMOSS--VLABench`

### Checkpoints & assets [#checkpoints--assets]

* This benchmark is primarily rollout- or official-result-driven; see catalog `requires` and the task YAML for policy/simulator checkpoints.

### Key environment variables [#key-environment-variables]

* `WORLDFOUNDRY_VLABENCH_ROOT`
* `WORLDFOUNDRY_VLABENCH_RESULTS_PATH`

## Run Evaluation [#run-evaluation]

After preparing the assets and candidate artifacts listed on this page, use the public run entry points below.

### Set Candidate Artifacts [#set-candidate-artifacts]

```bash
export WORLDFOUNDRY_GENERATED_ARTIFACT_DIR=/path/to/generated/artifacts
```

### Import Official Results [#import-official-results]

```bash
worldfoundry-eval zoo benchmark-run \
  --benchmark-id vlabench \
  --mode normalizer \
  --official-results-path /path/to/official/results-or-report \
  --generated-artifact-dir "${WORLDFOUNDRY_GENERATED_ARTIFACT_DIR}" \
  --output-dir tmp/vlabench/normalizer \
  --json
```

### Score Generated Artifacts [#score-generated-artifacts]

```bash
worldfoundry-eval zoo benchmark-run \
  --benchmark-id vlabench \
  --mode official-run \
  --generated-artifact-dir "${WORLDFOUNDRY_GENERATED_ARTIFACT_DIR}" \
  --output-dir tmp/vlabench/official-run \
  --json
```

### Direct In-Tree Runner [#direct-in-tree-runner]

```bash
worldfoundry-eval zoo benchmark-run --benchmark-id vlabench --mode official-validation --official-results-path <official_results.json> --output-dir <out> --json
```

## Requirements [#requirements]

### Inputs [#inputs]

* Official simulator runtime
* benchmark assets
* policy checkpoint
* official result dump

### Environment [#environment]

* `WORLDFOUNDRY_VLABENCH_RESULTS_PATH`

### Outputs [#outputs]

* `scorecard.json`
* `raw_results.jsonl`
* `evaluation/summary.json`

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