Skip to content

Benchmark

MTEB(eng, 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
36
Scores recorded
36
Sources
1
IQ weight
1.04

Score distribution

46.1%
Mean
45.9%
Median
5.2%
Std dev
41.3%
10th percentile
55.1%
90th percentile
35.9%
Min
35.9%58.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.04 for embedding models, measured over 36 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1NV-Embed-v2Clustering· MTEB Leaderboard data58.5%
2BGE En IclClustering· MTEB Leaderboard data57.9%
3Qwen3 Embedding 8BClustering· MTEB Leaderboard data56.3%
4Qwen3 Embedding 4BClustering· MTEB Leaderboard data55.1%
5BGE multilingual gemma2Clustering· MTEB Leaderboard data54.7%
6E5 Mistral 7B InstructClustering· MTEB Leaderboard data50.3%
7Qwen3 Embedding 0.6BClustering· MTEB Leaderboard data49.9%
8Text Embedding 3 LargeClustering· MTEB Leaderboard data49%
9Text Embedding 004Clustering· MTEB Leaderboard data47.5%
10Text Embedding 005Clustering· MTEB Leaderboard data47.4%
11Multilingual E5 Large InstructClustering· MTEB Leaderboard data47.1%
12bge-large-enClustering· MTEB Leaderboard data47%
13Gte LargeClustering· MTEB Leaderboard data46.8%
14UAE-Large-V1Clustering· MTEB Leaderboard data46.7%
15Text Embedding 3 SmallClustering· MTEB Leaderboard data46.6%
16Gte BaseClustering· MTEB Leaderboard data46.2%
17BGE Large En V1.5Clustering· MTEB Leaderboard data46.1%
18Text Embedding Ada 002Clustering· MTEB Leaderboard data45.9%
19BGE Base En V1.5Clustering· MTEB Leaderboard data45.8%
20Jina Embeddings v3Clustering· MTEB Leaderboard data45.2%
21E5 Large V2Clustering· MTEB Leaderboard data44.3%
22Granite Embedding Small English R2Clustering· MTEB Leaderboard data44.1%
23E5 Base V2Clustering· MTEB Leaderboard data44.1%
24nomic-embed-text-v1.5Clustering· MTEB Leaderboard data43.9%
25nomic-embed-text-v1Clustering· MTEB Leaderboard data43.9%
26bge-small-en-v1.5Clustering· MTEB Leaderboard data43.8%
27All Mpnet Base V2Clustering· MTEB Leaderboard data43.6%
28All MiniLM L6 V2Clustering· MTEB Leaderboard data41.9%
29All MiniLM L12 V2Clustering· MTEB Leaderboard data41.8%
30Jina Embeddings v2 Base ENClustering· MTEB Leaderboard data41.7%
31snowflake-arctic-embed-l-v2.0Clustering· MTEB Leaderboard data41.6%
32Multilingual E5 LargeClustering· MTEB Leaderboard data41.3%
33snowflake-arctic-embed-m-v2.0Clustering· MTEB Leaderboard data41%
34Granite Embedding 278m MultilingualClustering· MTEB Leaderboard data39.2%
35Paraphrase Multilingual MPNet v2Clustering· MTEB Leaderboard data38.5%
36Jina Embedding B EN v1Clustering· MTEB Leaderboard data35.9%