SeaEval - Fundamental NLP Tasks - RTE (Zero-Shot) — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 32B Instruct | 90.97 |
| 2 | Qwen 2.5 72B Instruct | 90.25 |
| 3 | Qwen 2.5 14B Instruct | 86.64 |
| 4 | Qwen 2.5 7B Instruct | 85.92 |
| 5 | Llama 3.1 70B Instruct | 84.84 |
| 6 | Qwen 2 72B Instruct | 84.48 |
| 7 | Qwen 2 7B Instruct | 82.31 |
| 8 | Sailor2-8B-Chat | 81.23 |
| 9 | Llama 3 70B Instruct | 80.87 |
| 10 | SeaLLMs-v3-7B-Chat | 78.7 |
| 11 | Qwen 2.5 3B Instruct | 77.98 |
| 12 | Gemma 2 9B (IT) | 74.73 |
| 13 | Gemma 2 2B (IT) | 72.92 |
| 14 | Qwen 2.5 1.5B Instruct | 70.4 |
| 15 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 68.59 |
Interactive version: theaggregate.ai/benchmark?slug=seaeval-fundamental-nlp-tasks-rte-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.