---
title: MiKASA-Robo
description: "ICLR 2026 tabletop memory suite: 32 RL tasks, later 90 language-conditioned VLA tasks. WorldFoundry imports official results; it does not run ManiSkill."
---

MiKASA-Robo is the tabletop half of **MIKASA** (Memory-Intensive Skills Assessment Suite for Agents) from AIRI, MIPT, and HSE. The ICLR 2026 paper is **32** ManiSkill RL tasks in **12** groups across four memory types: **object**, **spatial**, **sequential**, and **capacity** (occlusion recall, prior layouts, long action orders). The later **MIKASA-Robo-VLA** release grows to **90** language-conditioned tasks (**10** memory types), Short / Medium / Long horizons (25–2160 steps), and **22,500** oracle trajectories. Those two releases are **not** interchangeable.

Paper: [Memory, Benchmark & Robots](https://arxiv.org/abs/2502.10550) ([OpenReview](https://openreview.net/forum?id=9cLPurIZMj)). VLA docs: [mikasarobo.github.io](https://mikasarobo.github.io/). Code: [CognitiveAISystems/MIKASA-Robo](https://github.com/CognitiveAISystems/MIKASA-Robo).

WorldFoundry **imports** official result exports. It does **not** run ManiSkill, collect PPO/oracle trajectories, or pick RL vs VLA for you — declare that on the export.

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

## How to run

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

Optional `WORLDFOUNDRY_MIKASA_ROOT` points at a local runtime/repo. Default import path: `WORLDFOUNDRY_MIKASA_RESULTS_PATH`. Manifest: `worldfoundry/data/benchmarks/tasks/external/mikasa.yaml`. Outputs: `scorecard.json` (primary `success_rate`), `raw_results.jsonl`, `evaluation/summary.json`.

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