Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Science on GPQA Diamond (Acc, AMSR)

63.1Accuracy

Dense

12.6625.75538.8551.945Jul 29, 2025
Updated 19d ago

Evaluation Results

MethodLinks
2025.07
63.10
2025.07
61.10
2025.07
60.715.54
2025.07
58.118.12
2025.07
56.113.8
2025.07
53.518.1
2025.07
5317.5
2025.07
520
2025.07
50.817.6
2025.07
40.916.5
2025.07
33.318.9
2025.07
31.217.8
2025.07
29.38.2
2025.07
25.37.29
2025.07
14.64.27