SeaEval - Multilingual Reasoning - C-Eval (Zero-Shot): leaderboard
Metric: Accuracy (%). Source: huggingface.co. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 83.25 |
| 2 | Qwen 2 72B Instruct | 83.13 |
| 3 | Qwen 2.5 32B Instruct | 82.63 |
| 4 | Qwen 2.5 14B Instruct | 78.39 |
| 5 | SeaLLMs-v3-7B-Chat | 76.59 |
| 6 | Qwen 2 7B Instruct | 76.15 |
| 7 | Qwen 2.5 7B Instruct | 74.6 |
| 8 | Llama 3.1 70B Cpt Sea Lionv3 Instruct | 67.62 |
| 9 | Llama 3.1 70B Instruct | 66.13 |
| 10 | Qwen 2.5 3B Instruct | 65.38 |
| 11 | Llama 3 70B Instruct | 62.2 |
| 12 | Qwen 2.5 1.5B Instruct | 59.71 |
| 13 | Sailor2-8B-Chat | 59.46 |
| 14 | gemma2-9B-cpt-sea-lionv3-instruct | 57.22 |
| 15 | Gemma 2 9B (IT) | 55.23 |
Interactive version: theaggregate.ai/benchmark?slug=seaeval-multilingual-reasoning-c-eval-zero-shot · How It Works · Data refreshed daily, snapshot 2026-09-05.