SeaEval - Fundamental NLP Tasks - WNLI (Zero-Shot) — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 25 models tracked.

Top models

#ModelScore
1Qwen 2 72B Instruct88.73
2Qwen 2.5 32B Instruct87.32
3Llama 3.1 70B Instruct84.51
4Qwen 2.5 14B Instruct83.1
5Qwen 2.5 72B Instruct81.69
6Llama 3 70B Instruct78.87
7Gemma 2 9B (IT)77.46
8Qwen 2.5 7B Instruct76.06
9Qwen 2 7B Instruct71.83
10Qwen 2.5 3B Instruct64.79
11Llama 3.1 8B Instruct61.97
12SeaLLMs-v3-7B-Chat59.15
13Llama 3 8B Cpt Sea Lionv2.1 Instruct59.15
14Sailor2-8B-Chat54.93
15Llama 3 8B Instruct47.89

Interactive version: theaggregate.ai/benchmark?slug=seaeval-fundamental-nlp-tasks-wnli-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.