Jina Embeddings v2 Small En

Sep 2023
Jina Embeddings v2 Small En
Compare in
on
self-host onlyprice
26.5MTEB English v2
Sep 2023released
512Dimensions
8K tokensMax input
enLanguages

Rankingconfiguration unstated?

Full ranking →
1Octen Law 8B v176.82MoD Embedding74.63Litil Embed 0.6B73.34Dinghy Law 8B v172.65Nemotron 3 Embed 8B BF1671.8
···
157Arabert All Nli Triplet Matryoshka16.4158Jina Embeddings v2 Small En16159Rubert Base Cased Sentence15.1

Strong at

C
Embeddings & Retrieval#170 of 200

Alternatives?

ModelScoreCostPriceWhat changes
BetterBge Base En v1.5unstatedOpen×2.5—$0.005Max input: 8K tokens → 1K tokens+1lOpenOcten Law 8B v1×4.8——Memory (4-bit): 0 GB → 4.5 GBtBetterGte BaseunstatedOpen×2.5—$0.005Max input: 8K tokens → 1K tokens+1iBetterE5 Base v2unstatedOpen×2.5—$0.005Max input: 8K tokens → 1K tokens+1sBetterAll Mpnet Base v2unstatedOpen×2.3—$0.005Max input: 8K 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
LicenceApache-2.0free for commercial use

Benchmarks

in Embeddings & Retrieval
BEIR · Mean over tasks · English45.1#132
BEIR · Mean over task types · English45.1#132
BEIR · Retrieval · English45.1#132
JMTEB v2 · Clustering · Japanese9.6#159
JMTEB v2 · STS · Japanese53.3#138
LongEmbed · Mean over tasks · English5.5#169
LongEmbed · Mean over task types · English5.5#169
LongEmbed · Retrieval · English5.5#169
MTEB German v1 · Classification · German43.8#141
MTEB Europe v1 · Classification · multilingual (23 languages)43.8#127
MTEB Europe v1 · Instruction Reranking · English-0.7#80
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)2.4#176
MTEB Europe v1 · Pair Classification · multilingual (11 languages)74.9#149
MTEB Europe v1 · Clustering · multilingual (29 languages)20.3#166
MTEB Europe v1 · Retrieval · multilingual (25 languages)11.1#146
MTEB Europe v1 · Reranking · multilingual (12 languages)26.8#180
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)15.7#174
MTEB French v1 · Pair Classification · French49.9#180
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)1.2#154
MTEB Indic v1 · STS · multilingual (13 languages)6.3#123
MTEB Indic v1 · Reranking · multilingual (2 languages)35#173
MTEB Indic v1 · Retrieval · multilingual (14 languages)2.4#152
MTEB Indic v1 · Pair Classification · Hindi51.5#165
MTEB Indic v1 · Clustering · multilingual (21 languages)2.5#183
MTEB Dutch v1 · Reranking · Dutch42.5#206
MTEB Portuguese v1 · Multilabel Classification · Portuguese18#143
MTEB Law v1 · Mean over tasks · multilingual (3 languages)16#158
MTEB Law v1 · Mean over task types · multilingual (3 languages)16#158
MTEB Law v1 · Retrieval · multilingual (3 languages)16#158
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)21#125
MTEB Medical v1 · Mean over task types · multilingual (10 languages)18.4#126
MTEB Medical v1 · Retrieval · multilingual (10 languages)21.5#125
MTEB Medical v1 · Clustering · English24.1#219
MTEB Medical v1 · Reranking · Mandarin Chinese9.4#193
FollowIR · Mean over tasks · English-0.7#80
FollowIR · Mean over task types · English-0.7#80
FollowIR · Instruction Reranking · English-0.7#80
MTEB Spanish v1 · Pair Classification · Spanish52.3#172
MTEB Thai v1 · Pair Classification · Thai51.3#154
MTEB Thai v1 · Clustering · Thai2.4#187
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)18#175
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)49.4#188
MTEB Russian v1.1 · Pair Classification · Russian47.9#154
RTEB Finnish beta · Mean over tasks · English45.8#80
RTEB French beta · Mean over tasks · multilingual (2 languages)8.3#167
RTEB German beta · Mean over tasks · German2.9#217
RTEB Health beta · Mean over tasks · multilingual (4 languages)32.5#78
RTEB Law beta · Mean over tasks · multilingual (4 languages)21.6#196
MTEB multilingual v2 · Clustering · multilingual (208 languages)27.3#156
MTEB multilingual v2 · Classification · multilingual (119 languages)43.8#143
MTEB multilingual v2 · Retrieval · multilingual (120 languages)13.5#168
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)11.9#171
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)12.6#174
MTEB multilingual v2 · Instruction Reranking · English-0.7#80
MTEB multilingual v2 · Reranking · multilingual (22 languages)27.6#179
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)70.8#157

FAQ

How much does Jina Embeddings v2 Small En cost?›

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

Is Jina Embeddings v2 Small En free to use?›

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

How good is Jina Embeddings v2 Small En, and what is it best at?›

#170 of 200 in Embeddings.

Is Jina Embeddings v2 Small En better than Arabert All Nli Triplet Matryoshka?›

On MTEB Law v1 · Mean over tasks · multilingual (3 languages), Jina Embeddings v2 Small En scores 16 vs 16.4 — about the same.

When was Jina Embeddings v2 Small En released? Is there a newer version?›

Released 2023-09-27. It is the current version in its line.

Is Jina Embeddings v2 Small En open source?›

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

Can I run Jina Embeddings v2 Small En locally?›

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

What is Jina Embeddings v2 Small En?›

Jina Embeddings v2 Small En is an open-weights embeddings & retrieval model from Jina AI, released Sep 2023.

What are the best alternatives to Jina Embeddings v2 Small En?›

Bge Base En v1.5 (better), Octen Law 8B v1 (open weights).