HUG-VIS - Emotion Recognition (Video + Audio + Text): leaderboard
Metric: Seven-class emotion recognition accuracy (%; 8,400 seated half-body clips of 30 professional actors, each performing the same 280 emotion-action-prompt assignments in a controlled Mandarin studio, 1,200 clips per emotion; zero-shot; input: video, audio, text). Source: arxiv.org. Saturation forecast: Around December 2026. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen2.5-Omni-7B | 83.79 |
| 2 | MiniCPM-o-4.5 | 77.3 |
Interactive version: theaggregate.ai/benchmark?slug=hug-vis-emotion-recognition-video-plus-audio-plus-text · How It Works · Data refreshed daily, snapshot 2026-09-29.