STEM Reasoning on MATH (pass@1)
44.8pass@1Qwen3
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3Size=4B, Type=Base, Prompt Setting=4-shot2025.12 | 44.8 | |
| Youtu-LLMSize=2B, Type=Base, Prompt Setting=4-shot2025.12 | 44.4 | |
| SmolLM3Size=3B, Type=Base, Prompt Setting=4-shot2025.12 | 40.8 | |
| Qwen3Size=1.7B, Type=Base, Prompt Setting=4-shot2025.12 | 28.1 | |
| Gemma3Size=4B, Type=Base, Prompt Setting=4-shot2025.12 | 24.4 | |
| Llama3.1Size=8B, Type=Base, Prompt Setting=4-shot2025.12 | 21.5 |