Jina Embedding B En v1

Jul 2023
Jina Embedding B En v1
Compare in
on
$0.05per 1M tokens
46.4MTEB English v2
Jul 2023released
1K tokensMax input
enLanguages

Rankingconfiguration unstated?

Full ranking →
1Uibk Embedding v191.52GTE ModernColBERT v190.63UltraBERT Embed 150M79.34Voyage Multilingual 279.25Voyage Law 278.8
···
90Mmlw Roberta Large37.491Jina Embedding B En v137.292Sentence Croissant Alpha v0.437.2

Strong at

C
Embeddings & Retrieval#125 of 200

Alternatives?

ModelScoreCostPriceWhat changes
VCheaperVoyage 3 Liteunstated+92%-60%$0.02Languages: not verifiediBetterE5 Base v2unstatedOpen+10%-90%$0.005DOpenUibk Embedding v1unstated×2.5——Max input: 1K tokens → 0K tokensCheaperJina Embeddings v5 Text NanounstatedOpen+71%-60%$0.02Languages: not verifiedCheaperBge M3unstatedOpen+60%-80%$0.01Languages: not verified+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
LicenceApache-2.0free for commercial use

Benchmarks

in Embeddings & Retrieval
BEIR · Mean over tasks · English44#134
BEIR · Mean over task types · English44#134
BEIR · Retrieval · English44#134
JMTEB v2 · STS · Japanese4#163
LongEmbed · Mean over tasks · English37.2#91
LongEmbed · Mean over task types · English37.2#91
LongEmbed · Retrieval · English37.2#91
MTEB German v1 · Clustering · German25.3#78
MTEB German v1 · Retrieval · German42.6#37
MTEB German v1 · STS · German59.3#51
MTEB German v1 · Pair Classification · German50.1#80
MTEB German v1 · Reranking · German36.5#57
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)46.8#101
MTEB Europe v1 · Instruction Reranking · English-4.3#196
MTEB Europe v1 · Pair Classification · multilingual (11 languages)80.5#108
MTEB French v1 · Retrieval · multilingual (2 languages)44.9#42
MTEB French v1 · Reranking · French73.4#47
MTEB French v1 · Clustering · French39.4#68
MTEB French v1 · Summarization · French30.2#69
MTEB French v1 · Pair Classification · French58.5#86
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)6.8#74
MTEB Indic v1 · Pair Classification · Hindi52.6#150
MTEB Indic v1 · STS · multilingual (13 languages)3#144
MTEB Korean v2 · Reranking · Korean5.8#130
MTEB Korean v2 · Classification · Korean20.8#119
MTEB Dutch v1 · Mean over tasks · Dutch36.3#101
MTEB Dutch v1 · Mean over task types · Dutch44.7#100
MTEB Dutch v1 · Classification · Dutch44.8#103
MTEB Dutch v1 · Clustering · Dutch17#109
MTEB Dutch v1 · Retrieval · Dutch32#94
MTEB Dutch v1 · STS · Dutch57.2#100
MTEB Dutch v1 · Multilabel Classification · Dutch27.4#104
MTEB Dutch v1 · Pair Classification · Dutch61.6#91
MTEB Dutch v1 · Reranking · Dutch73#167
MTEB Polish v1 · Pair Classification · Polish60.8#128
MTEB Polish v1 · STS · Polish49.8#103
MTEB Portuguese v1 · Multilabel Classification · Portuguese15.5#181
MTEB Law v1 · Mean over tasks · multilingual (3 languages)40.8#78
MTEB Law v1 · Mean over task types · multilingual (3 languages)40.8#78
MTEB Law v1 · Retrieval · multilingual (3 languages)40.8#78
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)33.1#92
MTEB Medical v1 · Mean over task types · multilingual (10 languages)25.9#103
MTEB Medical v1 · Retrieval · multilingual (10 languages)36#91
MTEB Medical v1 · Clustering · English31.2#186
MTEB Medical v1 · Reranking · Mandarin Chinese10.4#180
FollowIR · Mean over tasks · English-4.3#196
FollowIR · Mean over task types · English-4.3#196
FollowIR · Instruction Reranking · English-4.3#196
MTEB Spanish v1 · Pair Classification · Spanish57.3#134
MTEB Spanish v1 · Clustering · Spanish39.1#91
MTEB Spanish v1 · Reranking · Spanish35.5#87
MTEB Thai v1 · Pair Classification · Thai51.6#140
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)27.8#129
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)57.2#156
MTEB Russian v1.1 · STS · Russian36.3#115
MTEB Russian v1.1 · Reranking · Russian19.8#121
MTEB Russian v1.1 · Clustering · Russian15.6#123
MTEB Russian v1.1 · Pair Classification · Russian48.8#141
MTEB Russian v1.1 · Retrieval · Russian4.7#78
MTEB Russian v1.1 · Multilabel Classification · Russian26.5#106
RTEB French beta · Mean over tasks · multilingual (2 languages)26.9#106
RTEB German beta · Mean over tasks · German29.9#100
RTEB Law beta · Mean over tasks · multilingual (4 languages)34.4#104
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)71.6#131
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)26.3#114
MTEB multilingual v2 · Instruction Reranking · English-4.3#196

FAQ

How much does Jina Embedding B En v1 cost?›

$0.05 per 1M tokens.

Is Jina Embedding B En v1 free to use?›

There is no free API; it costs $0.05 per 1M tokens. The weights are open, so you can run it yourself.

How good is Jina Embedding B En v1, and what is it best at?›

#125 of 200 in Embeddings.

Is Jina Embedding B En v1 better than Mmlw Roberta Large?›

On LongEmbed · Retrieval · English, Jina Embedding B En v1 scores 37.2 vs 37.4 — about the same.

When was Jina Embedding B En v1 released? Is there a newer version?›

Released 2023-07-07. It is the current version in its line.

Is Jina Embedding B En v1 open source?›

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

Can I run Jina Embedding B En v1 locally?›

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

What is Jina Embedding B En v1?›

Jina Embedding B En v1 is an open-weights embeddings & retrieval model from Jina AI, released Jul 2023.

What are the best alternatives to Jina Embedding B En v1?›

Voyage 3 Lite (cheaper), E5 Base v2 (better), Uibk Embedding v1 (open weights).