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

Science Question Answering on GPQA (Accuracy, Memory, and Time)

0.21Memory Ratio

AC

0.17840.39170.6050.8183Jan 29, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
0.212.1165.03
2026.01
0.250.3736.5
2026.01
0.250.4555.53
2026.01
0.262.0835.92
2026.01
0.270.5139.17
2026.01
0.270.4665.17
2026.01
0.312.1838.12
2026.01
0.322.3836.45
2026.01
0.321.9765.12
2026.01
0.362.2739.05
2026.01
0.363.9264.21
2026.01
0.372.9936.24
2026.01
0.382.738.79
2026.01
0.431.6755.42
2026.01
0.493.264.87
2026.01
0.524.1734.49
2026.01
0.533.4139.19
2026.01
0.542.6254.67
2026.01
0.64.0352.99
2026.01
0.653.853.13
2026.01
1139.49
2026.01
1136.66
2026.01
1155.29
2026.01
1165.33