SeaEval - Fundamental NLP Tasks - C3 (Zero-Shot) — leaderboard

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

Top models

#ModelScore
1Qwen 2 72B Instruct96.11
2Llama 3.1 70B Instruct96.04
3Qwen 2.5 32B Instruct96.04
4Qwen 2.5 72B Instruct95.96
5Llama 3 70B Instruct95.21
6Qwen 2.5 14B Instruct95.03
7Qwen 2 7B Instruct92.45
8Gemma 2 9B (IT)92.22
9SeaLLMs-v3-7B-Chat91.44
10Qwen 2.5 7B Instruct91.21
11Sailor2-8B-Chat89.6
12Llama 3.1 8B Instruct88.15
13Llama 3 8B Cpt Sea Lionv2.1 Instruct86.76
14Qwen 2.5 3B Instruct86.69
15Llama 3 8B Instruct85.15

Interactive version: theaggregate.ai/benchmark?slug=seaeval-fundamental-nlp-tasks-c3-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.