HELM MedHELM - SHC-Proxy: leaderboard

HELM MedHELM Stanford Health Care surrogate/proxy decision clinical task (exact match).

Metric: EM. Source: crfm.stanford.edu. 13 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.0 Flash74.67#342
2DeepSeek R174.33#251
3Claude 3.5 Sonnet (20241022)73.67#291
4Gemini 2.5 Pro (Preview 05-06)73.67#147
5GPT-4o Mini (2024-07-18)73.33#558
6GPT-4o (2024-05-13)72#325
7Gemini 1.5 Pro (001)71.33#444
8GPT-5 Mini (2025-08-07)69.67#172
9O3 Mini (2025-01-31)69.67#240
10O4 Mini (2025-04-16)69#179
11Llama 3.3 70B Instruct68.33#534
12GPT-568#98
13Claude 3.7 Sonnet (20250219)62#201

Interactive version: theaggregate.ai/benchmark?slug=helm-medhelm-shc-proxy · How It Works · Data refreshed daily, snapshot 2026-10-20.