Skip to content

Benchmark

MTEB(Indic, 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
34
Scores recorded
34
Sources
1
IQ weight
0.54

Score distribution

24.7%
Mean
31.3%
Median
18.7%
Std dev
2.9%
10th percentile
48.2%
90th percentile
1.7%
Min
1.7%55.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.54 for embedding models, measured over 34 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 data55.7%
2Multilingual E5 Large InstructClustering· MTEB Leaderboard data51.7%
3Qwen3 Embedding 4BClustering· MTEB Leaderboard data49.7%
4Jina Embeddings v5 Omni SmallClustering· MTEB Leaderboard data48.2%
5Jina Embeddings v5 Text SmallClustering· MTEB Leaderboard data48.2%
6Gemini Embedding 001Clustering· MTEB Leaderboard data47.2%
7Jina Embeddings v5 Omni NanoClustering· MTEB Leaderboard data44.7%
8Jina Embeddings v5 Text NanoClustering· MTEB Leaderboard data44.7%
9Jina Embeddings v3Clustering· MTEB Leaderboard data40.5%
10Qwen3 Embedding 0.6BClustering· MTEB Leaderboard data39.4%
11Text Multilingual Embedding 002Clustering· MTEB Leaderboard data39.2%
12BGE multilingual gemma2Clustering· MTEB Leaderboard data34.6%
13Granite Embedding 278m MultilingualClustering· MTEB Leaderboard data33.2%
14E5 Mistral 7B InstructClustering· MTEB Leaderboard data32.7%
15embeddinggemma-2Clustering· MTEB Leaderboard data32.2%
16Paraphrase Multilingual MPNet v2Clustering· MTEB Leaderboard data32.1%
17snowflake-arctic-embed-l-v2.0Clustering· MTEB Leaderboard data31.3%
18BGE M3Clustering· MTEB Leaderboard data30%
19Text Embedding 3 LargeClustering· MTEB Leaderboard data23.2%
20snowflake-arctic-embed-m-v2.0Clustering· MTEB Leaderboard data22.7%
21NV-Embed-v2Clustering· MTEB Leaderboard data13.1%
22Text Embedding 3 SmallClustering· MTEB Leaderboard data7.2%
23E5 Base V2Clustering· MTEB Leaderboard data4.1%
24E5 Large V2Clustering· MTEB Leaderboard data4%
25nomic-embed-text-v1.5Clustering· MTEB Leaderboard data3.8%
26nomic-embed-text-v1Clustering· MTEB Leaderboard data3.6%
27UAE-Large-V1Clustering· MTEB Leaderboard data3.1%
28All MiniLM L6 V2Clustering· MTEB Leaderboard data3.1%
29bge-small-en-v1.5Clustering· MTEB Leaderboard data3.1%
30BGE Base En V1.5Clustering· MTEB Leaderboard data3.1%
31BGE Large En V1.5Clustering· MTEB Leaderboard data2.9%
32Gte LargeClustering· MTEB Leaderboard data2.6%
33Gte BaseClustering· MTEB Leaderboard data2.1%
34Jina Embeddings v2 Base ENClustering· MTEB Leaderboard data1.7%