Skip to content

Benchmark

MedHELM: MedQA - EM

Science benchmarks measure how well models answer science questions. Scores are percentages. Higher is better.

Models measured
13
Scores recorded
13
Sources
1
IQ weight
1.12

Score distribution

87.3%
Mean
86.5%
Median
6.8%
Std dev
76.9%
10th percentile
95.6%
90th percentile
75%
Min
75%96.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.12 for chat models, measured over 13 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Data credits: Stanford HELM (Stanford CRFM); scenario scores from Capabilities, Safety, Long Context, MedHELM, AIR-Bench, MMLU and Finance