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

Scientific Reasoning on GPQA Diamond (Pass@1, Token count)

66.8Pass@1 Accuracy

AlphaOne

44.4450.24556.0561.855Apr 3, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.04
66.85,591
2026.04
66.82,477
2026.04
66.54,151
2026.04
66.26,151
2026.04
66.23,563
2026.04
664,362
2026.04
65.72,918
2026.04
65.35,475
2026.04
65.34,322
2026.04
65.24,001
2026.04
654,101
2026.04
64.82,247
2026.04
64.86,274
2026.04
64.63,544
2026.04
64.51,713
2026.04
64.55,881
2026.04
63.54,026
2026.04
63.14,502
2026.04
61.83,497
2026.04
61.36,771
2026.04
61.33,119
2026.04
60.86,027
2026.04
60.43,706
2026.04
60.43,596
2026.04
60.47,123
2026.04
60.34,048
2026.04
60.13,309
2026.04
59.86,002
2026.04
59.63,046
2026.04
59.43,139
2026.04
59.13,678
2026.04
59.17,472
2026.04
58.94,611
2026.04
58.86,638
2026.04
58.43,007
2026.04
56.92,926
2026.04
51.24,109
2026.04
50.58,591
2026.04
50.33,209
2026.04
49.88,890
2026.04
49.53,107
2026.04
49.28,016
2026.04
48.83,635
2026.04
47.83,593
2026.04
47.68,569
2026.04
47.34,423
2026.04
473,390
2026.04
473,624
2026.04
46.83,729
2026.04
46.68,783
2026.04
46.38,341
2026.04
46.14,410
2026.04
45.54,152
2026.04
45.33,299