EmoTrans - Emotion Transition Reasoning: leaderboard
EmoTrans (2026) emotion transition reasoning: the model explains why a person's emotion changed or stayed the same in a video clip, and a GPT-5 judge checks agreement with the reference explanation; LLM-Score (%) over 724 questions, higher is better.
Metric: LLM-Score (%) on Emotion Transition Reasoning (724 questions: a GPT-5 judge checks whether the model's explanation of why the emotion changed or stayed matches the reference explanation), on EmoTrans, built from 1,000 manually annotated real-world video clips (OpenHumanVid and CH-SIMS v2) over 12 scenarios with one or more target persons; zero-shot, video with its text (and audio where the model accepts it); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5.4 | 85.7 | #135 | |
| 2 | Gemini 2.5 Flash | Non-reasoning | 82.7 | #463 |
| 3 | Gemini 3 Pro | 82.3 | #145 | |
| 4 | Gemini 3 Flash | Non-reasoning | 82.3 | #175 |
| 5 | Seed 2.0 Lite | 81.4 | #300 | |
| 6 | GLM-4.5V | 79.5 | #754 | |
| 7 | GLM-4.6V | 79 | #710 | |
| 8 | Qwen 3 VL 32B Instruct | 78.9 | #606 | |
| 9 | Qwen 3.5 27B | 77.9 | #375 | |
| 10 | Qwen 3.5 122B A10B | 77.3 | #346 | |
| 11 | Qwen 2.5 VL 72B Instruct | 76.4 | #841 | |
| 12 | Qwen 3 Omni 30B A3B | 75.5 | #847 | |
| 13 | Qwen 3 VL 8B | 75.4 | #917 | |
| 14 | InternVL3.5-8B | 70.2 | #1604 | |
| 15 | InternVL3.5-14B | 65.8 | #1504 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=emotrans-emotion-transition-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.