SeaEval - Cultural Reasoning - SG-Eval v2 MCQ (Zero-Shot) — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 87.64 |
| 2 | Qwen 2.5 72B Instruct | 86.18 |
| 3 | Qwen 2 72B Instruct | 85.82 |
| 4 | Qwen 2.5 32B Instruct | 84.36 |
| 5 | Llama 3 70B Instruct | 83.82 |
| 6 | Qwen 2.5 14B Instruct | 83.45 |
| 7 | Llama 3.1 8B Instruct | 81.27 |
| 8 | Gemma 2 9B (IT) | 80.36 |
| 9 | Llama 3 8B Instruct | 79.09 |
| 10 | SeaLLMs-v3-7B-Chat | 79.09 |
| 11 | Qwen 2 7B Instruct | 78.73 |
| 12 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 78.36 |
| 13 | Qwen 2.5 7B Instruct | 78 |
| 14 | Sailor2-8B-Chat | 73.09 |
| 15 | Qwen 2.5 3B Instruct | 72 |
Interactive version: theaggregate.ai/benchmark?slug=seaeval-cultural-reasoning-sg-eval-v2-mcq-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.