RoboProcessBench - Phase Recognition: leaderboard

Metric: Accuracy (%) on 1,274 questions asking which coarse process phase a manipulation is in, from a single frame (4 options, chance 25.0); 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 2035. 14 models tracked.

Top models

#ModelScore
1InternVL3.5-8B37.4
2Claude Sonnet 4.631.9
3InternVL3-8B31.3
4GPT-5.4 Mini30.9
5GPT-4o29
6Qwen 3 VL 32B28.4
7Qwen 2.5 VL 7B Instruct26.6
8Claude Haiku 4.524.8
9InternVL3-38B24.6
10GLM-4.6V21.1

Interactive version: theaggregate.ai/benchmark?slug=roboprocessbench-phase-recognition · How It Works · Data refreshed daily, snapshot 2026-09-29.