---
title: RoboMME
description: "ICML 2026 memory-augmented manipulation: 16 tasks in Counting, Permanence, Reference, and Imitation. WorldFoundry imports suite success; it does not run the sim."
---

RoboMME (University of Michigan with Stanford and Figure AI; ICML 2026) is a standardized bench for **memory-augmented** VLAs. **16** long-horizon tasks sit in four cognitive suites — **Counting** (temporal: exact event counts), **Permanence** (spatial: occlusion and swaps), **Reference** (object identity from an earlier cue), **Imitation** (procedural: copy a demonstrated trajectory) — four tasks each. The release includes **1,600** demonstrations and **770k** training timesteps, plus **14** memory-augmented variants on a **π0.5** backbone. The paper reports per-suite success, not a single pick-and-place bit.

Paper: [RoboMME](https://arxiv.org/abs/2603.04639). Project: [robomme.github.io](https://robomme.github.io/). Code: [RoboMME/robomme_policy_learning](https://github.com/RoboMME/robomme_policy_learning).

WorldFoundry **imports** official rollout exports and normalizes overall plus per-suite rates. It does **not** run the RoboMME simulator or those π0.5 variants. Name the suite (`counting` / `permanence` / `reference` / `imitation`) on the export.

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

## How to run

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

Optional `WORLDFOUNDRY_ROBOMME_ROOT` points at a local runtime/repo. Default import path: `WORLDFOUNDRY_ROBOMME_RESULTS_PATH`. Manifest: `worldfoundry/data/benchmarks/tasks/external/robomme.yaml`. Outputs: `scorecard.json` (primary `success_rate`; also `counting_success_rate`, `permanence_success_rate`, `reference_success_rate`, `imitation_success_rate`), `raw_results.jsonl`, `evaluation/summary.json`.

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