RoboProcessBench - Next Primitive Prediction: leaderboard
Metric: Accuracy (%) on 200 questions asking which low-level primitive should happen next (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 2031. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-38B | 63.5 |
| 2 | GLM-4.6V | 60 |
| 3 | Claude Haiku 4.5 | 56 |
| 4 | InternVL3.5-8B | 53.5 |
| 5 | Qwen 3 VL 32B | 47 |
| 6 | InternVL3-8B | 46.5 |
| 7 | GPT-5.4 Mini | 45.5 |
| 8 | Claude Sonnet 4.6 | 44.8 |
| 9 | GPT-4o | 44.5 |
| 10 | Qwen 2.5 VL 7B Instruct | 33.1 |
Interactive version: theaggregate.ai/benchmark?slug=roboprocessbench-next-primitive-prediction · How It Works · Data refreshed daily, snapshot 2026-09-29.