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

Scientific Knowledge and Reasoning on HealthBench

63.3Score

GPT-5.2 High

27.31636.6584655.342Jun 30, 2026
Updated 23d ago

Evaluation Results

MethodLinks
63.3
2026.06
62.5
57.7
2026.06
51.6
51.2
37.9
36.3
30
28.7