Skip to content

Benchmark

MTEB(cmn, v1)

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
1.19

Score distribution

64.9%
Mean
63.6%
Median
5.8%
Std dev
59.9%
10th percentile
72.3%
90th percentile
57.8%
Min
57.8%73.9%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.19 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