List your company

Science Reasoning on GPQA Diamond (Accuracy (%))

75.7AccuracyDeepSeek-R1
61.55665.22868.972.572Aug 13, 2025

Evaluation results

MethodLinks
DeepSeek-R12025.0875.7
DS-32B-MathDomain=Math2025.0868.69
NTele-R1-32B2025.0867.68
DS-32B-CodeDomain=Code2025.0865.15
QwQ-32B2025.0863.6
DS-32B2025.0862.1