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

GPQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Question AnsweringGPQA
Accuracy84.2
258
Science ReasoningGPQA
Accuracy95.1
243
Graduate-level Question AnsweringGPQA
Accuracy96.9
215
ReasoningGPQA Diamond
Accuracy91.9
185
Scientific Question AnsweringGPQA Diamond
Accuracy84.4
123
Graduate-level Science ReasoningGPQA
Accuracy82.34
121
Scientific ReasoningGPQA Main
Accuracy43.08
101
General ReasoningSuper GPQA
Accuracy71.1
99
Question AnsweringGPQA Diamond
Accuracy68.4
97
ReasoningGPQA
Accuracy59.4
88
Science Question AnsweringGPQA
pass@1 Accuracy87.6
85
Scientific ReasoningGPQA Diamond (test)
Accuracy99.37
82
Scientific ReasoningGPQA-D
Accuracy (%)66.2
77
Scientific ReasoningGPQA
Accuracy70.2
75
Science ReasoningGPQA
Accuracy (GPQA)59.22
72
Graduate-level Science Question AnsweringGPQA
Accuracy (GPQA)61.95
72
Science Question AnsweringGPQA
Accuracy63.51
69
Science Question AnsweringGPQA
Accuracy55.8
69
Scientific ReasoningGPQA Diamond
Score73.2
68
Scientific ReasoningGPQA Diamond
Pass@1 Accuracy78.66
67
Question AnsweringGPQA (test)
Accuracy70.4
65
Closed-ended reasoningGPQA Diamond (test)
Accuracy65.9
63
Scientific ReasoningGPQA Diamond
Accuracy80.13
62
Question AnsweringGPQA Diamond
Accuracy84
61
Science Question AnsweringGPQA main (test)
Exact Match Accuracy40.85
60
Showing 25 of 475 rows
...