STEM Knowledge and Reasoning on MMLU STEM
82.3Pass@1Qwen2.5-Math-72B-Instruct
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen2.5-Math-72B-InstructReasoning Mode=Tool-Integrated Reasoning, Shots=5-shot2024.09 | 82.3 | 85 | 90 | — | |
| Qwen2.5-Math-72B-InstructReasoning Mode=Chain-of-Thought, Shots=5-shot2024.09 | 80.8 | 84.9 | 80.1 | — | |
| GPT-4o-2024-08-06Reasoning Mode=Chain-of-Thought, Shots=5-shot2024.09 | 64.2 | — | — | — |