AdaptMMBench - Key Step Coverage: leaderboard
Metric: Key step coverage (%): how far the adaptive-mode reasoning chain progresses along the human-verified key reasoning steps, judged by GPT-5 (open models only; closed models' traces are not available); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B Instruct | 84.83 | #264 |
| 2 | Qwen 3 VL 32B Instruct | 83.79 | #276 |
| 3 | Qwen 3 VL 8B Instruct | 78.4 | #401 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=adaptmmbench-key-step-coverage · How It Works · Data refreshed daily, snapshot 2026-10-11.