EgoCoT-Bench - High-level Grounded Reasoning: leaderboard
Metric: Answer accuracy (%, four-way multiple choice, strict exact match) of zero-shot MLLMs on egocentric manipulation videos of EgoCoT-Bench (351 clips, 3,172 human-verified questions built from spatio-temporal scene graphs), High-level Grounded Reasoning group (progress reasoning, hand-object grounded chain of thought and goal-oriented object tracking); higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 397B A17B | 68.08 |
| 2 | Qwen 3.5 Plus | 67.64 |
| 3 | Qwen 3.5 27B | 65.3 |
| 4 | GPT-5.1 | 65.15 |
| 5 | GPT-5.2 | 64.86 |
| 6 | Qwen 3.5 122B A10B | 64.28 |
| 7 | Qwen 3 VL 235B A22B | 60.18 |
Interactive version: theaggregate.ai/benchmark?slug=egocot-bench-high-level-grounded-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-07.