Skip to content

Benchmark

MTEB(Europe, v1) - STS

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

Models measured
23
Scores recorded
23
Sources
1
IQ weight
0.55

Score distribution

72.8%
Mean
74.3%
Median
6.5%
Std dev
64.2%
10th percentile
81.7%
90th percentile
62.7%
Min
62.7%81.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.55 for embedding models, measured over 23 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard