List your company

LLM-as-a-judge alignment on Clinician Spine cohort (val R1)

78.43-Way AgreementMedGemma-27B-text-it
41.79251.29660.870.304May 10, 2026

Evaluation results

MethodLinks
MedGemma-27B-text-it2026.0578.40.03
Qwen3-32B2026.0564.90.05
GPT-4.12026.0559.50.04
GPT-4.1-mini2026.0555.40.03
GPT-52026.0543.20.02