Commonsense Knowledge on MMLU-ProX-Zh
45.2AccuracyQwen3
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3Size=4B, Type=Base, Prompt Setting=5-shot2025.12 | 45.2 | |
| Youtu-LLMSize=2B, Type=Base, Prompt Setting=5-shot2025.12 | 40.7 | |
| Qwen3Size=1.7B, Type=Base, Prompt Setting=5-shot2025.12 | 32.5 | |
| SmolLM3Size=3B, Type=Base, Prompt Setting=5-shot2025.12 | 26.7 | |
| Llama3.1Size=8B, Type=Base, Prompt Setting=5-shot2025.12 | 25.4 | |
| Gemma3Size=4B, Type=Base, Prompt Setting=5-shot2025.12 | 24.2 |