RoboStressBench - Planning: leaderboard
Metric: Accuracy (%) on the 821 planning multiple-choice questions, exact match on the predicted option, greedy decoding with at most 64 new tokens (API models: temperature 0, 64 output tokens), one unified instruction template; images of embodied scenes under physical visual stress (material, viewpoint, lighting, geometry); higher is better. Source: arxiv.org. Saturation forecast: Around March 2028. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 35B A3B | 92.9 |
| 2 | Qwen 3.6 35B A3B | 81.1 |
| 3 | Qwen 3.5 27B | 77 |
| 4 | Qwen 3.5 4B | 68.8 |
| 5 | Qwen 3.6 27B | 68.8 |
| 6 | Qwen 3.5 9B | 65.5 |
| 7 | GPT-5.5 | 57 |
| 8 | Gemini 3.1 Pro (Preview) | 56.3 |
| 9 | Molmo2-8B | 54.3 |
| 10 | InternVL3.5-8B | 50.3 |
Interactive version: theaggregate.ai/benchmark?slug=robostressbench-planning · How It Works · Data refreshed daily, snapshot 2026-09-29.