Skip to content

Benchmark

MTEB(rus, v1) - Clustering

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

Models measured
31
Scores recorded
31
Sources
1
IQ weight
0.83

Score distribution

39.7%
Mean
50.9%
Median
20.5%
Std dev
15.8%
10th percentile
63.1%
90th percentile
13.3%
Min
13.3%66.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.83 for embedding models, measured over 31 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Qwen3 Embedding 8BClustering· MTEB Leaderboard data66.7%
2Qwen3 Embedding 4BClustering· MTEB Leaderboard data65.1%
3E5 Mistral 7B InstructClustering· MTEB Leaderboard data64.2%
4Multilingual E5 Large InstructClustering· MTEB Leaderboard data63.1%
5Jina Embeddings v5 Omni SmallClustering· MTEB Leaderboard data61.4%
6Jina Embeddings v5 Text SmallClustering· MTEB Leaderboard data61.3%
7Jina Embeddings v3Clustering· MTEB Leaderboard data60.9%
8Qwen3 Embedding 0.6BClustering· MTEB Leaderboard data60.5%
9Jina Embeddings v5 Omni NanoClustering· MTEB Leaderboard data60%
10Text Embedding 3 LargeClustering· MTEB Leaderboard data57.8%
11snowflake-arctic-embed-l-v2.0Clustering· MTEB Leaderboard data53.9%
12Text Embedding 3 SmallClustering· MTEB Leaderboard data52.7%
13BGE M3Clustering· MTEB Leaderboard data52.4%
14Multilingual E5 LargeClustering· MTEB Leaderboard data52.2%
15NV-Embed-v2Clustering· MTEB Leaderboard data51.7%
16Granite Embedding 278m MultilingualClustering· MTEB Leaderboard data50.9%
17Paraphrase Multilingual MPNet v2Clustering· MTEB Leaderboard data49.2%
18Gte LargeClustering· MTEB Leaderboard data22.3%
19E5 Large V2Clustering· MTEB Leaderboard data22%
20E5 Base V2Clustering· MTEB Leaderboard data19.2%
21UAE-Large-V1Clustering· MTEB Leaderboard data19.1%
22BGE Large En V1.5Clustering· MTEB Leaderboard data19%
23nomic-embed-text-v1.5Clustering· MTEB Leaderboard data17.9%
24nomic-embed-text-v1Clustering· MTEB Leaderboard data17.8%
25BGE Base En V1.5Clustering· MTEB Leaderboard data17.6%
26Gte BaseClustering· MTEB Leaderboard data17%
27bge-small-en-v1.5Clustering· MTEB Leaderboard data15.8%
28All Mpnet Base V2Clustering· MTEB Leaderboard data15.8%
29Jina Embedding B EN v1Clustering· MTEB Leaderboard data15.6%
30All MiniLM L12 V2Clustering· MTEB Leaderboard data13.8%
31All MiniLM L6 V2Clustering· MTEB Leaderboard data13.3%