Bge Small En v1.5

Sep 2023
Bge Small En v1.5
Compare in
on
$0.02per 1M tokens
53.9MTEB English v2
Sep 2023released
384Dimensions
1K tokensMax input
enLanguages

Rankingconfiguration unstated?

Full ranking →
1F2LLM v2 8B78.62F2LLM v2 14B78.43F2LLM v2 4B76.54Text Embedding 3 Large75.95ICT TIME And Querit Embedding v175.2
···
29Bge Base En69.230Bge Small En v1.569.231E5 Base69.1

Strong at

B
Embeddings & Retrieval#70 of 200

Alternatives?

ModelScoreCostPriceWhat changes
CheaperBge Large En v1.5unstatedOpen+3%-50%$0.01Memory (4-bit): 0 GB → 0.2 GBBetterText Embedding 3 Smallunstated+6%same$0.02Languages: not verifiedcOpenF2LLM v2 8Bunstated+14%——Memory (4-bit): 0 GB → 4.5 GBCheaperBge Base En v1.5unstatedOpen+2%-75%$0.005Memory (4-bit): 0 GB → 0.1 GBsCheaperAll Mpnet Base v2unstatedOpensame-75%$0.005Max input: 1K tokens → 0K tokens+1

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
LicenceMITfree for commercial use

History

2 versions
68.86969.2Release dateChemTEB (task avg)Bge Small Env1.5
68.86969.2Release dateChemTEB (task avg)Bge Small Env1.5

Benchmarks

in Embeddings & Retrieval
BEIR · Mean over tasks · English51.7#83
BEIR · Mean over task types · English51.7#83
BEIR · Retrieval · English51.7#83
ChemTEB · Mean over tasks · multilingual (13 languages)69.2#30
ChemTEB · Mean over task types · multilingual (13 languages)47.6#37
ChemTEB · Classification · English81.7#23
ChemTEB · Retrieval · English66.7#38
ChemTEB · Pair Classification · multilingual (13 languages)55.4#31
ChemTEB · Clustering · English34.3#33
ChemTEB · Bitext Mining · English0.1#35
MTEB Persian v2 · Reranking · Persian30.4#107
JMTEB v2 · Clustering · Japanese17.6#109
JMTEB v2 · STS · Japanese62.6#94
LongEmbed · Mean over tasks · English32.1#116
LongEmbed · Mean over task types · English32.1#116
LongEmbed · Retrieval · English32.1#116
MTEB German v1 · Mean over tasks · German39.4#62
MTEB German v1 · Mean over task types · German38.7#59
MTEB German v1 · Clustering · German24.1#91
MTEB German v1 · Retrieval · German36#63
MTEB German v1 · Classification · German47.5#118
MTEB German v1 · STS · German49.9#82
MTEB German v1 · Pair Classification · German48.8#141
MTEB German v1 · Reranking · German26#79
MTEB Europe v1 · Mean over tasks · multilingual (29 languages)46.1#47
MTEB Europe v1 · Mean over task types · multilingual (29 languages)40.9#45
MTEB Europe v1 · Retrieval · multilingual (25 languages)38.5#85
MTEB Europe v1 · STS · multilingual (8 languages)66.8#46
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)33.2#132
MTEB Europe v1 · Classification · multilingual (23 languages)50.2#87
MTEB Europe v1 · Clustering · multilingual (29 languages)35#113
MTEB Europe v1 · Instruction Reranking · English-1.9#121
MTEB Europe v1 · Pair Classification · multilingual (11 languages)83#88
MTEB Europe v1 · Reranking · multilingual (12 languages)50.2#107
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)13.5#82
MTEB French v1 · Mean over tasks · multilingual (2 languages)48.5#56
MTEB French v1 · Mean over task types · multilingual (2 languages)49.8#58
MTEB French v1 · STS · French68.1#69
MTEB French v1 · Retrieval · multilingual (2 languages)36.4#75
MTEB French v1 · Reranking · French65.6#84
MTEB French v1 · Clustering · French42.8#39
MTEB French v1 · Classification · French52.2#87
MTEB French v1 · Summarization · French29.6#99
MTEB French v1 · Pair Classification · French54.3#160
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)32.1#93
MTEB Indic v1 · Mean over task types · multilingual (25 languages)22.5#97
MTEB Indic v1 · Classification · multilingual (8 languages)42.9#97
MTEB Indic v1 · Retrieval · multilingual (14 languages)11.1#117
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)2.6#127
MTEB Indic v1 · Reranking · multilingual (2 languages)43.5#146
MTEB Indic v1 · Pair Classification · Hindi50.8#169
MTEB Indic v1 · STS · multilingual (13 languages)3.2#141
MTEB Indic v1 · Clustering · multilingual (21 languages)3.1#158
MTEB Korean v2 · STS · Korean41.4#55
MTEB Korean v2 · Reranking · Korean13.5#88
MTEB Korean v2 · Classification · Korean23.1#78
MTEB Dutch v1 · Mean over tasks · Dutch39.8#85
MTEB Dutch v1 · Mean over task types · Dutch47.4#85
MTEB Dutch v1 · Classification · Dutch46.6#93
MTEB Dutch v1 · Clustering · Dutch21.6#94
MTEB Dutch v1 · Retrieval · Dutch38.1#78
MTEB Dutch v1 · STS · Dutch59.2#89
MTEB Dutch v1 · Multilabel Classification · Dutch29.4#93
MTEB Dutch v1 · Pair Classification · Dutch58.8#116
MTEB Dutch v1 · Reranking · Dutch78.3#132
MTEB Polish v1 · Pair Classification · Polish65.2#92
MTEB Polish v1 · Classification · Polish40.9#96
MTEB Polish v1 · STS · Polish57#66
MTEB Portuguese v1 · Multilabel Classification · Portuguese18.5#129
MTEB Law v1 · Mean over tasks · multilingual (3 languages)38.2#99
MTEB Law v1 · Mean over task types · multilingual (3 languages)38.2#99
MTEB Law v1 · Retrieval · multilingual (3 languages)38.2#99
MTEB Medical v1 · Clustering · English35.6#115
MTEB Medical v1 · Reranking · Mandarin Chinese20.9#132
FollowIR · Mean over tasks · English-1.9#121
FollowIR · Mean over task types · English-1.9#121
FollowIR · Instruction Reranking · English-1.9#121
MTEB Spanish v1 · Pair Classification · Spanish59.5#109
MTEB Spanish v1 · STS · multilingual (2 languages)63.3#53
MTEB Spanish v1 · Clustering · Spanish39.9#86
MTEB Spanish v1 · Reranking · Spanish36#84
MTEB Thai v1 · Clustering · Thai3.7#143
MTEB Thai v1 · Pair Classification · Thai51.1#164
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)32.2#108
MTEB Scandinavian v1 · Classification · multilingual (6 languages)48.4#103
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)62.4#114
MTEB Russian v1.1 · Mean over tasks · Russian26.7#74
MTEB Russian v1.1 · Mean over task types · Russian27.6#70
MTEB Russian v1.1 · Classification · Russian29.6#112
MTEB Russian v1.1 · STS · Russian40#106
MTEB Russian v1.1 · Reranking · Russian27.1#98
MTEB Russian v1.1 · Clustering · Russian15.8#120
MTEB Russian v1.1 · Pair Classification · Russian44.2#195
MTEB Russian v1.1 · Retrieval · Russian9.2#67
MTEB Russian v1.1 · Multilabel Classification · Russian27.1#95
RTEB Finnish beta · Mean over tasks · English50.9#70
RTEB French beta · Mean over tasks · multilingual (2 languages)17.3#140
RTEB German beta · Mean over tasks · German11.7#172
RTEB Health beta · Mean over tasks · multilingual (4 languages)39.2#70
RTEB Law beta · Mean over tasks · multilingual (4 languages)29.9#143
MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages)43.8#73
MTEB multilingual v2 · Mean over task types · multilingual (1037 languages)37.7#72
MTEB multilingual v2 · Classification · multilingual (119 languages)47.4#113
MTEB multilingual v2 · Clustering · multilingual (208 languages)39.9#100
MTEB multilingual v2 · STS · multilingual (35 languages)59.7#80
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)71.6#133
MTEB multilingual v2 · Retrieval · multilingual (120 languages)36.3#121
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)23.6#127
MTEB multilingual v2 · Instruction Reranking · English-1.9#121
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)17#94
MTEB multilingual v2 · Reranking · multilingual (22 languages)45.9#109

FAQ

How much does Bge Small En v1.5 cost?›

$0.02 per 1M tokens.

Is Bge Small En v1.5 free to use?›

There is no free API; it costs $0.02 per 1M tokens. The weights are open, so you can run it yourself.

How good is Bge Small En v1.5, and what is it best at?›

#70 of 200 in Embeddings.

Is Bge Small En v1.5 better than Bge Base En?›

On ChemTEB · Mean over tasks · multilingual (13 languages), Bge Small En v1.5 scores 69.2 vs 69.2 — about the same.

When was Bge Small En v1.5 released? Is there a newer version?›

Released 2023-09-12. It is the current version in its line.

Is Bge Small En v1.5 open source?›

Yes, the weights are open (MIT licence).

Can I run Bge Small En v1.5 locally?›

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

What is Bge Small En v1.5?›

Bge Small En v1.5 is an open-weights embeddings & retrieval model from BAAI, released Sep 2023.

What are the best alternatives to Bge Small En v1.5?›

Bge Large En v1.5 (cheaper), Text Embedding 3 Small (better), F2LLM v2 8B (open weights).