MOV-Bench - Hypothetical Reasoning: leaderboard
The 60 hypothetical-reasoning questions of MOV-Bench (Xu et al., 2026), asking what would follow under a supposed change from audio-visual evidence spread across several segments of a FineVideo clip. Accuracy in percent when the omni-modal model answers directly from the whole video; higher is better.
Metric: Accuracy (%; 60 hypothetical-reasoning questions; direct inference on the whole video). Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3 Omni 30B A3B Instruct | 35 | #902 |
| 2 | Qwen2.5-Omni-7B | 25 | #1797 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=mov-bench-hypothetical-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-12.