ATOM-Bench - Dual-Arm Composition: leaderboard
Held-out compositional manipulation on a real Agilex Cobot Magic dual-arm robot from ATOM-Bench (Wu et al., 2026): 12 bimanual tasks that combine motor skills with two or more language constraints and add hand-role assignment and inter-arm coordination, none of them demonstrated during fine-tuning. The policy is fine-tuned only on the benchmark's dual-arm atomic tasks; success rate in percent over 10 fixed physical test seeds per task, macro-averaged over tasks; higher is better.
Metric: Success rate (%; held-out Composition Set, 12 tasks x 10 fixed physical test seeds on the Agilex Cobot Magic dual-arm track, macro-averaged over tasks; one policy fine-tuned on the 15 atomic tasks of the track). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Pi0.5 | 16.7 | |
| 2 | Motus | 7.5 | |
| 3 | GR00T-N1.6 | 1.7 | |
| 4 | LingBot-VLA | 0 | |
| 5 | SmolVLA | 0 |
Interactive version: theaggregate.ai/benchmark?slug=atom-bench-dual-arm-composition · How It Works · Data refreshed daily, snapshot 2026-11-12.