LongEmotion: leaderboard
LongEmotion — emotion understanding and support over long conversational contexts; overall score.
Metric: Overall Score (%). Source: longemotion.github.io. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3 | 63.42 | #321 |
| 2 | GPT-4o | 61.29 | #343 |
| 3 | GPT-4o Mini | 59.1 | #599 |
| 4 | Qwen 3 8B | 58.98 | #674 |
| 5 | Llama 3.1 8B Instruct | 55.85 | #1035 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=longemotion · How It Works · Data refreshed daily, snapshot 2026-10-20.