RoboProcessBench - Motion State: leaderboard
Metric: Accuracy (%) on 896 questions asking whether the manipulator is actively moving or effectively stationary (2 options, chance 50.0); zero-shot multiple-choice VQA on held-out robot manipulation recordings from GM-100, RH20T, REASSEMBLE and AIST-Bimanual (ProcessData-Eval), temperature 0.01; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 32B | 69.2 |
| 2 | GPT-4o | 68.4 |
| 3 | Claude Sonnet 4.6 | 68.1 |
| 4 | GPT-5.4 Mini | 67 |
| 5 | Claude Haiku 4.5 | 61.3 |
| 6 | InternVL3-38B | 58.1 |
| 7 | Qwen 2.5 VL 7B Instruct | 58 |
| 8 | InternVL3-8B | 54 |
| 9 | InternVL3.5-8B | 53.8 |
| 10 | GLM-4.6V | 48.9 |
Interactive version: theaggregate.ai/benchmark?slug=roboprocessbench-motion-state · How It Works · Data refreshed daily, snapshot 2026-09-29.