List your company

Physics Reasoning on Human-Adapted (Pedagogical)

61.4Accuracy (Textbooks)Claude-Opus-4.5
5.2419.8234.448.98Apr 21, 2026

Evaluation results

MethodLinks
Claude-Opus-4.5Model Category=Proprietary2026.0461.463.155.762
Gemini-2.5-flashModel Category=Proprietary2026.0459.26359.161.4
OSS-20bModel Category=Open-Weight2026.0442.546.744.345.1
Qwen3-4B-Thinking-2507Model Category=Open-Weight2026.0418.92415.720.5
Qwen3-4B-Instruct-2507Model Category=Open-Weight2026.0410.213511.5
DeepSeek-7BModel Category=Open-Weight2026.047.48.16.47.8