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

Medical Reasoning on HealthBench

70.41Accuracy

COTCAgent

20.791633.673346.55559.4367May 14, 2026May 15, 2026May 16, 2026May 17, 2026May 18, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
70.41-68.7
2026.05
69.58-66.8
2026.05
68.81-67.8
2026.05
68.59-67.4
2026.05
68.43-66.8
2026.05
68.19-65.8
2026.05
63.87-62.7
2026.05
63.76-62.5
2026.05
63.41-61.8
2026.05
62.87-60.6
2026.05
62.31-61.2
2026.05
62.18-60.9
2026.05
61.94-60.7
2026.05
61.76-59.7
2026.05
61.44-60.2
2026.05
61.22-60
2026.05
61.08-59.8
2026.05
60.71-59.2
2026.05
60.58-58.6
2026.05
60.54-59
2026.05
60.42-58.7
2026.05
59.71-58.3
2026.05
59.52-58.1
2026.05
59.38-57.9
2026.05
58.63-56.6
2026.05
56.07-54.1
2026.05
55.74-53.4
2026.05
55.41-53.6
2026.05
54.88-52.8
2026.05
53.96-51.7
2026.05
34--
2026.05
33.6--
2026.05
33--
2026.05
32.9--
2026.05
31.2--
2026.05
22.7--
-63.3-
2026.02
-60.1-
2026.02
-66.2-
2026.02
-65.1-