Skip to content

Benchmark

MTEB(Scandinavian, 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
33
Scores recorded
35
Sources
1
IQ weight
1.49

Score distribution

38.4%
Mean
36.1%
Median
9%
Std dev
27.8%
10th percentile
51.9%
90th percentile
14.9%
Min
14.9%55.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.49 for embedding models, measured over 33 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 data55.3%
2Qwen3 Embedding 4BClustering· MTEB Leaderboard data54.9%
3Jina Embeddings v5 Omni SmallClustering· MTEB Leaderboard data52%
4Jina Embeddings v5 Text SmallClustering· MTEB Leaderboard data51.9%
5E5 Mistral 7B InstructClustering· MTEB Leaderboard data51.4%
6Jina Embeddings v5 Omni NanoClustering· MTEB Leaderboard data51.2%
7Multilingual E5 Large InstructClustering· MTEB Leaderboard data49.8%
8Qwen3 Embedding 0.6BClustering· MTEB Leaderboard data48.1%
9Text Embedding 3 LargeClustering· MTEB Leaderboard data42.8%
10Text Embedding 3 LargeClustering, embed_dim: 512· MTEB Leaderboard data42.3%
11Multilingual E5 LargeClustering· MTEB Leaderboard data41.8%
12Text Embedding 3 SmallClustering, embed_dim: 512· MTEB Leaderboard data41.3%
13Text Embedding 3 SmallClustering· MTEB Leaderboard data40.8%
14Jina Embeddings v3Clustering· MTEB Leaderboard data40.7%
15BGE M3Clustering· MTEB Leaderboard data38.8%
16snowflake-arctic-embed-l-v2.0Clustering· MTEB Leaderboard data37.9%
17Granite Embedding 278m MultilingualClustering· MTEB Leaderboard data37.6%
18UAE-Large-V1Clustering· MTEB Leaderboard data36.1%
19Paraphrase Multilingual MPNet v2Clustering· MTEB Leaderboard data35.7%
20BGE Large En V1.5Clustering· MTEB Leaderboard data34.8%
21snowflake-arctic-embed-m-v2.0Clustering· MTEB Leaderboard data34.7%
22Gte LargeClustering· MTEB Leaderboard data34.4%
23nomic-embed-text-v1Clustering· MTEB Leaderboard data34%
24Gte BaseClustering· MTEB Leaderboard data33.9%
25BGE Base En V1.5Clustering· MTEB Leaderboard data33.3%
26NV-Embed-v2Clustering· MTEB Leaderboard data33%
27E5 Large V2Clustering· MTEB Leaderboard data32.9%
28nomic-embed-text-v1.5Clustering· MTEB Leaderboard data32.8%
29E5 Base V2Clustering· MTEB Leaderboard data32.7%
30All Mpnet Base V2Clustering· MTEB Leaderboard data32.3%
31bge-small-en-v1.5Clustering· MTEB Leaderboard data32.2%
32Jina Embedding B EN v1Clustering· MTEB Leaderboard data27.8%
33All MiniLM L6 V2Clustering· MTEB Leaderboard data26.6%
34All MiniLM L12 V2Clustering· MTEB Leaderboard data24.7%
35Jina Embeddings v2 Base ENClustering· MTEB Leaderboard data14.9%