STEM on GSM8K
80.8pass@1Qwen3
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3Size=4B, Type=Base, Prompt Setting=8-shot2025.12 | 80.8 | |
| Youtu-LLMSize=2B, Type=Base, Prompt Setting=8-shot2025.12 | 77.6 | |
| Qwen3Size=1.7B, Type=Base, Prompt Setting=8-shot2025.12 | 68.2 | |
| SmolLM3Size=3B, Type=Base, Prompt Setting=8-shot2025.12 | 67.3 | |
| Llama3.1Size=8B, Type=Base, Prompt Setting=8-shot2025.12 | 47.8 | |
| Gemma3Size=4B, Type=Base, Prompt Setting=8-shot2025.12 | 38.5 |