E3mo-Bench (Expressed) - Single-Video Perception (What-How): leaderboard
E3mo-Bench (E³mo-Bench, 2026) expressed-emotion subset, about the emotion expressed by the people in a video: what or how multiple-choice questions on one video about its emotion, valence, arousal, dominance or cause. Accuracy in percent; higher is better.
Metric: Accuracy (%; what or how multiple-choice questions about one video's emotion, valence, arousal, dominance or its cause, one per video among 2,598 single-video questions; 16 MLLMs on videos from eight emotion datasets, each assigned one affective perspective; subset: the emotion expressed by the people on screen). Source: arxiv.org. Saturation forecast: Around June 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 70.21 | #123 |
| 2 | Claude Sonnet 4.6 | 67.67 | #161 |
| 3 | InternVL3-8B | 64.51 | #1688 |
| 4 | Qwen2.5-Omni-7B | 61.05 | #1789 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=e3mo-bench-expressed-single-video-perception-what-how · How It Works · Data refreshed daily, snapshot 2026-11-13.