Physics Reasoning on QFT Synthetic Easy
89AccuracyGemini-2.5-flash
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gemini-2.5-flashModel Type=Proprietary2026.04 | 89 | 96.2 | |
| Claude-Opus-4.5Model Type=Proprietary2026.04 | 81.5 | 95 | |
| OSS-20bModel Type=Open-Weight2026.04 | 81.5 | 95 | |
| Qwen3-4B-Thinking-2507Model Type=Open-Weight2026.04 | 69.5 | 83.8 | |
| Qwen3-4B-Instruct-2507Model Type=Open-Weight2026.04 | 53.5 | 75 | |
| DeepSeek-R1-Distill-Qwen-7BModel Type=Open-Weight2026.04 | 40.2 | 67.5 |