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

Clinical Question Answering on HealthBench (offline)

31.79Accuracy

SkeMex

7.963614.149320.33526.5207Jun 8, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
31.79
2026.06
28.66
2026.06
28.28
2026.06
27.65
2026.06
27.57
2026.06
27.56
2026.06
27.5
2026.06
27.36
2026.06
27.23
2026.06
26.89
2026.06
26.89
2026.06
26.74
2026.06
26.6
2026.06
26.29
2026.06
26
2026.06
25.85
2026.06
25.44
2026.06
25.06
2026.06
24.54
2026.06
23.98
2026.06
23.76
2026.06
23.74
2026.06
23.63
2026.06
23.49
2026.06
23.42
2026.06
23.37
2026.06
23.26
2026.06
23.08
2026.06
22.84
2026.06
22.42
2026.06
21.97
2026.06
21.59
2026.06
21.53
2026.06
21.44
2026.06
19.06
2026.06
14.83
2026.06
9.11
2026.06
8.88