List your company

Physics Reasoning on QFT Synthetic Hard

44.5AccuracyClaude-Opus-4.5
-1.7810.23522.2534.265Apr 21, 2026

Evaluation results

MethodLinks
Claude-Opus-4.5Model Type=Proprietary2026.0444.562.5
Gemini-2.5-flashModel Type=Proprietary2026.0430.257.5
OSS-20bModel Type=Open-Weight2026.0422.253.8
Qwen3-4B-Thinking-2507Model Type=Open-Weight2026.043.27.5
Qwen3-4B-Instruct-2507Model Type=Open-Weight2026.042.56.2
DeepSeek-R1-Distill-Qwen-7BModel Type=Open-Weight2026.0400