L

Bilingual Embedding Small

Jul 2024
LBilingual Embedding Small
Compare in
on
self-host onlyprice
41.7MTEB English v2
Jul 2024released
384Dimensions
1K tokensMax input
en, frLanguages

Rankingconfiguration unstated?

Full ranking →
1F2LLM v2 8B88.52F2LLM v2 14B88.53F2LLM v2 4B88.44F2LLM v2 1.7B88.15F2LLM v2 0.6B87.3
···
38KaLM Embedding Multilingual Mini v165.439Bilingual Embedding Small65.440Multilingual E5 Base65.3

Strong at

C
Embeddings & Retrieval#140 of 200

Alternatives?

ModelScoreCostPriceWhat changes
iBetterMultilingual E5 Large InstructunstatedOpen+6%—$0.01Memory (4-bit): 0.1 GB → 0.3 GBcOpenF2LLM v2 8Bunstated+35%——Memory (4-bit): 0.1 GB → 4.5 GBBetterBge M3unstatedOpen+5%—$0.01Languages: not verified+1iBetterMultilingual E5 LargeunstatedOpen+3%—$0.01Memory (4-bit): 0.1 GB → 0.3 GBBetterJina Embeddings v5 Text NanounstatedOpen+31%—$0.02Languages: not verified

Run it yourself

estimate
Fits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use

Benchmarks

in Embeddings & Retrieval
MTEB Persian v2 · Reranking · Persian66.4#41
JMTEB v2 · Clustering · Japanese50.7#29
JMTEB v2 · STS · Japanese79.9#44
LongEmbed · Mean over tasks · English37.1#93
LongEmbed · Mean over task types · English37.1#93
LongEmbed · Retrieval · English37.1#93
MTEB German v1 · Mean over tasks · German52.1#23
MTEB German v1 · Mean over task types · German51#26
MTEB German v1 · Clustering · German31.8#36
MTEB German v1 · Retrieval · German46.3#30
MTEB German v1 · Classification · German65.4#39
MTEB German v1 · STS · German70.8#21
MTEB German v1 · Pair Classification · German51.4#62
MTEB German v1 · Reranking · German40.3#51
MTEB Europe v1 · Retrieval · multilingual (25 languages)48.7#42
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)84.9#55
MTEB Europe v1 · Classification · multilingual (23 languages)57#38
MTEB Europe v1 · Clustering · multilingual (29 languages)39.5#66
MTEB Europe v1 · Instruction Reranking · English-3.4#178
MTEB Europe v1 · Pair Classification · multilingual (11 languages)87.1#40
MTEB Europe v1 · Reranking · multilingual (12 languages)55.1#73
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)15.8#58
MTEB French v1 · Retrieval · multilingual (2 languages)48.7#37
MTEB French v1 · Reranking · French75.1#43
MTEB French v1 · Clustering · French46.6#19
MTEB French v1 · Classification · French66.4#34
MTEB French v1 · Summarization · French32.3#5
MTEB French v1 · Pair Classification · French58.2#92
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)65#18
MTEB Indic v1 · Mean over task types · multilingual (25 languages)64.5#17
MTEB Indic v1 · Classification · multilingual (8 languages)63.5#21
MTEB Indic v1 · Retrieval · multilingual (14 languages)77.6#29
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)71.1#21
MTEB Indic v1 · Reranking · multilingual (2 languages)83.7#54
MTEB Indic v1 · Pair Classification · Hindi71.2#57
MTEB Indic v1 · STS · multilingual (13 languages)47.8#35
MTEB Indic v1 · Clustering · multilingual (21 languages)36.7#42
MTEB Korean v2 · Reranking · Korean49.2#46
MTEB Korean v2 · Classification · Korean56.1#39
MTEB Dutch v1 · Mean over tasks · Dutch51.9#49
MTEB Dutch v1 · Mean over task types · Dutch58.5#47
MTEB Dutch v1 · Classification · Dutch59.3#45
MTEB Dutch v1 · Clustering · Dutch31.8#47
MTEB Dutch v1 · Retrieval · Dutch52.4#56
MTEB Dutch v1 · STS · Dutch76.6#46
MTEB Dutch v1 · Multilabel Classification · Dutch34.7#46
MTEB Dutch v1 · Pair Classification · Dutch70#61
MTEB Dutch v1 · Reranking · Dutch85.1#86
MTEB Polish v1 · Pair Classification · Polish79.4#52
MTEB Polish v1 · Classification · Polish55.8#46
MTEB Polish v1 · STS · Polish66.2#33
MTEB Portuguese v1 · Multilabel Classification · Portuguese17.7#149
MTEB Law v1 · Mean over tasks · multilingual (3 languages)49.4#51
MTEB Law v1 · Mean over task types · multilingual (3 languages)49.4#51
MTEB Law v1 · Retrieval · multilingual (3 languages)49.4#51
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)46.7#45
MTEB Medical v1 · Mean over task types · multilingual (10 languages)49.3#44
MTEB Medical v1 · Retrieval · multilingual (10 languages)47.5#49
MTEB Medical v1 · Clustering · English33.2#161
MTEB Medical v1 · Reranking · Mandarin Chinese67.2#81
FollowIR · Mean over tasks · English-3.4#178
FollowIR · Mean over task types · English-3.4#178
FollowIR · Instruction Reranking · English-3.4#178
MTEB Spanish v1 · Pair Classification · Spanish63.3#71
MTEB Spanish v1 · Clustering · Spanish44.6#41
MTEB Spanish v1 · Reranking · Spanish50.7#51
MTEB Thai v1 · Clustering · Thai45.2#49
MTEB Thai v1 · Pair Classification · Thai69.1#60
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)41#44
MTEB Scandinavian v1 · Classification · multilingual (6 languages)57.9#52
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)66.6#78
MTEB Russian v1.1 · Mean over tasks · Russian56.4#37
MTEB Russian v1.1 · Mean over task types · Russian55.3#36
MTEB Russian v1.1 · Classification · Russian56.5#62
MTEB Russian v1.1 · STS · Russian71.3#41
MTEB Russian v1.1 · Reranking · Russian58.4#52
MTEB Russian v1.1 · Clustering · Russian53.3#49
MTEB Russian v1.1 · Pair Classification · Russian55.4#88
MTEB Russian v1.1 · Retrieval · Russian59.4#35
MTEB Russian v1.1 · Multilabel Classification · Russian32.7#59
RTEB French beta · Mean over tasks · multilingual (2 languages)26.8#107
RTEB German beta · Mean over tasks · German44.1#79
RTEB Law beta · Mean over tasks · multilingual (4 languages)40.1#81
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)77.1#65
MTEB multilingual v2 · Retrieval · multilingual (120 languages)49.5#80
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)69.5#41
MTEB multilingual v2 · Instruction Reranking · English-3.4#178
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)19.2#67
MTEB multilingual v2 · Reranking · multilingual (22 languages)59.3#72

FAQ

How much does Bilingual Embedding Small cost?›

There is no per-unit API price: self-host only.

Is Bilingual Embedding Small free to use?›

There is no free API. The weights are open, so you can run it yourself.

How good is Bilingual Embedding Small, and what is it best at?›

#140 of 200 in Embeddings.

Is Bilingual Embedding Small better than KaLM Embedding Multilingual Mini v1?›

On MTEB German v1 · Classification · German, Bilingual Embedding Small scores 65.4 vs 65.4 — about the same.

When was Bilingual Embedding Small released? Is there a newer version?›

Released 2024-07-17. It is the current version in its line.

Is Bilingual Embedding Small open source?›

Yes, the weights are open (Apache-2.0 licence).

Can I run Bilingual Embedding Small locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

What is Bilingual Embedding Small?›

Bilingual Embedding Small is an open-weights embeddings & retrieval model from Lajavaness, released Jul 2024.

What are the best alternatives to Bilingual Embedding Small?›

Multilingual E5 Large Instruct (better), F2LLM v2 8B (open weights).