SeaEval - Emotion - SST-2 (Zero-Shot) — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 95.3 |
| 2 | Llama 3 70B Instruct | 94.95 |
| 3 | Qwen 2.5 32B Instruct | 94.72 |
| 4 | Sailor2-8B-Chat | 94.61 |
| 5 | SeaLLMs-v3-7B-Chat | 94.04 |
| 6 | Qwen 2 7B Instruct | 93.46 |
| 7 | Qwen 2 72B Instruct | 93.46 |
| 8 | Qwen 2.5 72B Instruct | 93.35 |
| 9 | Gemma 2 9B (IT) | 93.12 |
| 10 | Qwen 2.5 14B Instruct | 93.12 |
| 11 | Qwen 2.5 7B Instruct | 92.55 |
| 12 | Gemma 2 2B (IT) | 92.43 |
| 13 | Qwen 2.5 1.5B Instruct | 92.32 |
| 14 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 91.74 |
| 15 | Llama 3.1 8B Instruct | 89.56 |
Interactive version: theaggregate.ai/benchmark?slug=seaeval-emotion-sst-2-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.