Skip to content

Benchmark

MTEB(Europe, v1) - BitextMining

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

Models measured
37
Scores recorded
37
Sources
1
IQ weight
1.55

Score distribution

66.8%
Mean
83.7%
Median
26.3%
Std dev
33.2%
10th percentile
90.9%
90th percentile
1.3%
Min
1.3%98.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.55 for embedding models, measured over 37 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 2BitextMining· MTEB Leaderboard data98.5%
2Qwen3 Embedding 8BBitextMining· MTEB Leaderboard data93%
3Qwen3 Embedding 4BBitextMining· MTEB Leaderboard data92.2%
4Jina Embeddings v5 Omni SmallBitextMining· MTEB Leaderboard data90.9%
5Jina Embeddings v5 Text SmallBitextMining· MTEB Leaderboard data90.9%
6Multilingual E5 Large InstructBitextMining· MTEB Leaderboard data90.4%
7Jina Embeddings v5 Omni NanoBitextMining· MTEB Leaderboard data90.1%
8Jina Embeddings v5 Text NanoBitextMining· MTEB Leaderboard data90.1%
9Gemini Embedding 001BitextMining· MTEB Leaderboard data89.8%
10Multilingual E5 LargeBitextMining· MTEB Leaderboard data88%
11Text Embedding 3 LargeBitextMining· MTEB Leaderboard data88%
12Text Multilingual Embedding 002BitextMining· MTEB Leaderboard data87.3%
13Qwen3 Embedding 0.6BBitextMining· MTEB Leaderboard data86.8%
14Jina Embeddings v3BitextMining· MTEB Leaderboard data86.2%
15snowflake-arctic-embed-l-v2.0BitextMining· MTEB Leaderboard data85.9%
16NV-Embed-v2BitextMining· MTEB Leaderboard data85.7%
17E5 Mistral 7B InstructBitextMining· MTEB Leaderboard data85.5%
18embeddinggemma-2BitextMining· MTEB Leaderboard data84.2%
19Text Embedding 3 SmallBitextMining· MTEB Leaderboard data83.7%
20Granite Embedding 278m MultilingualBitextMining· MTEB Leaderboard data81.1%
21Paraphrase Multilingual MPNet v2BitextMining· MTEB Leaderboard data79.5%
22snowflake-arctic-embed-m-v2.0BitextMining· MTEB Leaderboard data76.3%
23E5 Large V2BitextMining· MTEB Leaderboard data53.8%
24Jina Embedding B EN v1BitextMining· MTEB Leaderboard data46.8%
25E5 Base V2BitextMining· MTEB Leaderboard data45.9%
26UAE-Large-V1BitextMining· MTEB Leaderboard data42.6%
27BGE Large En V1.5BitextMining· MTEB Leaderboard data41.8%
28Gte LargeBitextMining· MTEB Leaderboard data41.4%
29nomic-embed-text-v1BitextMining· MTEB Leaderboard data40.7%
30nomic-embed-text-v1.5BitextMining· MTEB Leaderboard data39.4%
31Gte BaseBitextMining· MTEB Leaderboard data36.8%
32BGE Base En V1.5BitextMining· MTEB Leaderboard data34.1%
33bge-small-en-v1.5BitextMining· MTEB Leaderboard data33.2%
34All MiniLM L12 V2BitextMining· MTEB Leaderboard data32.1%
35All Mpnet Base V2BitextMining· MTEB Leaderboard data29.9%
36All MiniLM L6 V2BitextMining· MTEB Leaderboard data27.2%
37Jina Embeddings v2 Base ENBitextMining· MTEB Leaderboard data1.3%