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

Scientific Reasoning on GPQA Diamond (Acc, Tok, CR, Cost)

65.25Accuracy

Full

14.0327.327540.62553.9225Apr 6, 2026Apr 15, 2026Apr 25, 2026May 4, 2026May 14, 2026May 23, 2026Jun 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
65.25---
2026.04
65.28,4471008,447
2026.04
65.28,09895.98,433
2026.06
61.68---
2026.06
59.09---
2026.04
58.36,35575.26,533
2026.06
57.39---
2026.06
56.76---
2026.06
56.19---
2026.04
563,76344.53,799
2026.04
55.89,5361009,536
2026.04
55.89,42498.89,825
2026.04
54.16,8657210,371
2026.04
53.83,128373,159
2026.04
53.27,3181007,318
2026.04
53.27,243997,558
2026.04
536,41267.26,496
2026.04
52.86,06463.66,144
2026.04
52.86,8049311,142
2026.04
51.75,88380.45,967
2026.04
50.95,634775,715
2026.06
49.94---
2026.06
49.05---
2026.04
496,45967.76,670
2026.04
496,33486.66,535
2026.06
48.48---
2026.06
47.98---
2026.04
47.59,05698.718,243
2026.04
47.49,1771009,177
2026.04
47.49,1781009,527
2026.05
47.22,870.9--
2026.04
478,03287.58,468
2026.04
46.97,89886.18,318
2026.04
45.77,94886.68,375
2026.04
44.78309.81,060
2026.05
44.13,492.9--
2026.06
43.94---
2026.05
40.84,295.6--
2026.05
40.36,491.1--
2026.04
39.19479.91,210
2026.05
38.35,225--
2026.06
37.94---
2026.05
37.95,431.5--
2026.06
36.87---
2026.05
35.27,288.7--
2026.04
33.91,23116.81,656
2026.04
33.32,62928.64,394
2026.05
25.43,259.4--
2026.05
255,324--
2026.05
24.44,013.8--
2026.05
23.96,566.8--
2026.05
235,285.2--
2026.05
21.85,136.9--
2026.05
168,514.3--