Skip to content

Benchmark

MTEB(cmn, v1) - Retrieval

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

Models measured
17
Scores recorded
17
Sources
1
IQ weight
1.43

Score distribution

53.2%
Mean
64.9%
Median
26.4%
Std dev
6%
10th percentile
76%
90th percentile
3.6%
Min
3.6%78.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.43 for embedding models, measured over 16 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard