Skip to content

Benchmark

R2MED

Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.

Models measured
7
Scores recorded
7
Sources
1
IQ weight
1.02

Score distribution

18.5%
Mean
17%
Median
6.4%
Std dev
14.9%
10th percentile
24.9%
90th percentile
8.7%
Min
8.7%30%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.02 for embedding models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard