Research

Health Optimization Bench

84.9

83.8

81.3

78.3

77.9

77.8

77.7

68.4

57.9

53.9

50.1

35.5

33.0

29.2

14.4

7.8

TM

Claude Fable 5.1

Claude Fable 5

Grok 4.6

Claude Opus 5

GPT-5.6 Sol (max)

Kimi K3

GPT-5.6 Sol (high)

Muse Spark

Gemini 3.6

Inkling

Claude Sonnet 5

MiniMax M3

MAI Thinking

GLM 5.2

Mistral Medium 3.5

Nemotron 3.5 Lightning

Health Optimization Bench ranks frontier language models on rubric-graded questions of current clinical evidence, authored and verified across independent model families.

The evaluation harness is open source, and the public task sample is published on Hugging Face.