Skip to content

Benchmark

MTEB(eng, v2) - Clustering

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

Models measured
39
Scores recorded
39
Sources
1
IQ weight
0.84

Score distribution

48.2%
Mean
47.7%
Median
5.8%
Std dev
43.5%
10th percentile
54.7%
90th percentile
25.4%
Min
25.4%59.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.84 for embedding models, measured over 39 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini Embedding 001Clustering· MTEB Leaderboard data59.4%
2Qwen3 Embedding 8BClustering· MTEB Leaderboard data58.6%
3Qwen3 Embedding 4BClustering· MTEB Leaderboard data57.5%
4Jina Embeddings v5 Omni SmallClustering· MTEB Leaderboard data54.7%
5Jina Embeddings v5 Text SmallClustering· MTEB Leaderboard data54.7%
6Qwen3 Embedding 0.6BClustering· MTEB Leaderboard data54.1%
7Jina Embeddings v5 Omni NanoClustering· MTEB Leaderboard data53.5%
8Jina Embeddings v5 Text NanoClustering· MTEB Leaderboard data53.5%
9Text Embedding 005Clustering· MTEB Leaderboard data51.9%
10embeddinggemma-2Clustering· MTEB Leaderboard data51.7%
11Text Embedding 004Clustering· MTEB Leaderboard data51.5%
12E5 Mistral 7B InstructClustering· MTEB Leaderboard data51.4%
13Text Multilingual Embedding 002Clustering· MTEB Leaderboard data50.4%
14Multilingual E5 Large InstructClustering· MTEB Leaderboard data49.9%
15Text Embedding 3 LargeClustering· MTEB Leaderboard data48.9%
16Gte LargeClustering· MTEB Leaderboard data48.2%
17BGE Large En V1.5Clustering· MTEB Leaderboard data48%
18UAE-Large-V1Clustering· MTEB Leaderboard data47.9%
19Gte BaseClustering· MTEB Leaderboard data47.7%
20NV-Embed-v2Clustering· MTEB Leaderboard data47.7%
21nomic-embed-text-v1.5Clustering· MTEB Leaderboard data47.6%
22Text Embedding 3 SmallClustering· MTEB Leaderboard data47.5%
23BGE Base En V1.5Clustering· MTEB Leaderboard data47.4%
24Jina Embeddings v3Clustering· MTEB Leaderboard data47.4%
25bge-small-en-v1.5Clustering· MTEB Leaderboard data47%
26nomic-embed-text-v1Clustering· MTEB Leaderboard data46.6%
27All Mpnet Base V2Clustering· MTEB Leaderboard data46.5%
28Granite Embedding Small English R2Clustering· MTEB Leaderboard data46.5%
29E5 Large V2Clustering· MTEB Leaderboard data45.2%
30E5 Base V2Clustering· MTEB Leaderboard data45.2%
31All MiniLM L6 V2Clustering· MTEB Leaderboard data44.9%
32All MiniLM L12 V2Clustering· MTEB Leaderboard data44.5%
33snowflake-arctic-embed-l-v2.0Clustering· MTEB Leaderboard data44.4%
34snowflake-arctic-embed-m-v2.0Clustering· MTEB Leaderboard data43.6%
35Paraphrase Multilingual MPNet v2Clustering· MTEB Leaderboard data43.5%
36Granite Embedding 278m MultilingualClustering· MTEB Leaderboard data43.3%
37Multilingual E5 LargeClustering· MTEB Leaderboard data42.1%
38Jina Embedding B EN v1Clustering· MTEB Leaderboard data40%
39Jina Embeddings v2 Base ENClustering· MTEB Leaderboard data25.4%