i

Multilingual E5 Large

Feb 2024
iMultilingual E5 Large
Compare in
on
$0.01per 1M tokens
51.5MTEB English v2
Feb 2024released
1,024Dimensions
1K tokensMax input
en, Esperanto, es +90Languages

Rankingconfiguration unstated?

Full ranking →
1E5 Large Matryoshka Sts Pt882Serafim 900M Portuguese Pt Sentence Encoder86.73Legal BERTimbau Sts Large Ma v386.54BERTimbau Large Matryoshka Sts Pt86.15ModBERTBr Matryoshka Sts Pt85.2
···
9Gte Multilingual Base80.410Multilingual E5 Large79.211Paraphrase Multilingual Mpnet Base v278.7

Strong at

B
Embeddings & Retrieval#94 of 200

Alternatives?

ModelScoreCostPriceWhat changes
CheaperEmbeddingGemma 300MunstatedOpen+2%-80%$0.002Languages: not verifiediBetterMultilingual E5 Large InstructunstatedOpen+4%same$0.01iOpenE5 Large Matryoshka Sts Pt+11%——Languages: loses en, Esperanto, es +89PBetterSerafim 900M Portuguese Pt Sentence EncoderOpen+9%——Max input: 1K tokens → 0K tokens+2rBetterLegal BERTimbau Sts Large Ma v3Open+9%——Languages: loses en, Esperanto, es +89

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

Benchmarks

in Embeddings & Retrieval
BEIR · Mean over tasks · English51.5#85
BEIR · Mean over task types · English51.5#85
BEIR · Retrieval · English51.5#85
BEIR-NL · Mean over tasks · Dutch45.9#3
BEIR-NL · Mean over task types · Dutch45.9#3
BEIR-NL · Retrieval · Dutch45.9#3
MTEB Mandarin Chinese v1 · Mean over tasks · Mandarin Chinese57.8#26
MTEB Mandarin Chinese v1 · Mean over task types · Mandarin Chinese57.9#26
MTEB Mandarin Chinese v1 · Retrieval · Mandarin Chinese63.7#59
MTEB Mandarin Chinese v1 · STS · Mandarin Chinese45.5#57
MTEB Mandarin Chinese v1 · Reranking · Mandarin Chinese57.5#58
MTEB Mandarin Chinese v1 · Clustering · Mandarin Chinese46.6#56
MTEB Mandarin Chinese v1 · Pair Classification · Mandarin Chinese64.2#59
MTEB Mandarin Chinese v1 · Classification · Mandarin Chinese69.8#28
ChemTEB · Mean over tasks · multilingual (13 languages)68.3#37
ChemTEB · Mean over task types · multilingual (13 languages)48.5#33
ChemTEB · Classification · English79.2#39
ChemTEB · Retrieval · English73#21
ChemTEB · Pair Classification · multilingual (13 languages)57#22
ChemTEB · Clustering · English32.9#38
ChemTEB · Bitext Mining · English0.2#23
MTEB Persian v2 · Mean over tasks · Persian64.3#13
MTEB Persian v2 · Mean over task types · Persian67.4#14
MTEB Persian v2 · STS · Persian79.3#9
MTEB Persian v2 · Clustering · Persian57.2#33
MTEB Persian v2 · Classification · Persian65.7#18
MTEB Persian v2 · Retrieval · Persian56#11
MTEB Persian v2 · Reranking · Persian74.3#3
MTEB Persian v2 · Bitext Mining · Persian56.6#21
MTEB Persian v2 · Pair Classification · Persian82.3#18
JMTEB v2 · STS · Japanese80.8#32
LongEmbed · Mean over tasks · English40.4#71
LongEmbed · Mean over task types · English40.4#71
LongEmbed · Retrieval · English40.4#71
MTEB German v1 · Mean over tasks · German54.9#17
MTEB German v1 · Mean over task types · German55#14
MTEB German v1 · Clustering · German33.5#28
MTEB German v1 · Retrieval · German50#23
MTEB German v1 · Classification · German67.5#33
MTEB German v1 · STS · German71.6#16
MTEB German v1 · Pair Classification · German54.1#31
MTEB German v1 · Reranking · German53.6#11
MTEB Europe v1 · STS · multilingual (8 languages)75.6#10
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)88#36
MTEB Europe v1 · Instruction Reranking · English-3.1#165
MTEB Europe v1 · Pair Classification · multilingual (11 languages)88.8#27
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)14.6#65
MTEB French v1 · Reranking · French77.4#34
MTEB French v1 · Summarization · French30.9#39
MTEB Indic v1 · Retrieval · multilingual (14 languages)82.6#14
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)77.7#6
MTEB Indic v1 · Pair Classification · Hindi75.1#35
MTEB Indic v1 · STS · multilingual (13 languages)43.9#42
MTEB Korean v2 · STS · Korean82.1#18
MTEB Korean v2 · Reranking · Korean55.3#30
MTEB Korean v2 · Classification · Korean60.8#20
MTEB Dutch v1 · Mean over tasks · Dutch54.7#35
MTEB Dutch v1 · Mean over task types · Dutch62.2#28
MTEB Dutch v1 · Classification · Dutch60.6#36
MTEB Dutch v1 · Clustering · Dutch29#54
MTEB Dutch v1 · Retrieval · Dutch59#30
MTEB Dutch v1 · STS · Dutch80.1#24
MTEB Dutch v1 · Multilabel Classification · Dutch35.9#38
MTEB Dutch v1 · Pair Classification · Dutch80.3#19
MTEB Dutch v1 · Reranking · Dutch90.3#31
MTEB Polish v1 · Pair Classification · Polish85.3#21
MTEB Polish v1 · Classification · Polish61#29
MTEB Portuguese v1 · Mean over tasks · Portuguese69.2#6
MTEB Portuguese v1 · Mean over task types · Portuguese63#7
MTEB Portuguese v1 · Classification · Portuguese59.3#9
MTEB Portuguese v1 · Retrieval · Portuguese74.5#7
MTEB Portuguese v1 · Multilabel Classification · Portuguese17.7#150
MTEB Portuguese v1 · STS · Portuguese79.2#10
MTEB Portuguese v1 · Reranking · Portuguese84.2#6
MTEB Law v1 · Mean over tasks · multilingual (3 languages)48.4#54
MTEB Law v1 · Mean over task types · multilingual (3 languages)48.4#54
MTEB Law v1 · Retrieval · multilingual (3 languages)48.4#54
MTEB Medical v1 · Clustering · English32.9#166
MTEB Medical v1 · Reranking · Mandarin Chinese66.8#83
FollowIR · Mean over tasks · English-3.1#165
FollowIR · Mean over task types · English-3.1#165
FollowIR · Instruction Reranking · English-3.1#165
MTEB Spanish v1 · Mean over tasks · multilingual (2 languages)64.7#15
MTEB Spanish v1 · Mean over task types · multilingual (2 languages)65.6#16
MTEB Spanish v1 · Pair Classification · Spanish65.2#63
MTEB Spanish v1 · STS · multilingual (2 languages)79.3#11
MTEB Spanish v1 · Retrieval · multilingual (2 languages)50.5#14
MTEB Spanish v1 · Clustering · Spanish46.6#27
MTEB Spanish v1 · Reranking · Spanish61.7#19
MTEB Spanish v1 · Classification · Spanish90.5#10
MTEB Thai v1 · Pair Classification · Thai71#53
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)41.8#32
MTEB Scandinavian v1 · Retrieval · multilingual (3 languages)64.5#25
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)69.1#55
VN-MTEB Vietnamese v1 · Mean over tasks · Vietnamese51.5#11
VN-MTEB Vietnamese v1 · Mean over task types · Vietnamese62.5#10
VN-MTEB Vietnamese v1 · Classification · Vietnamese63.1#16
VN-MTEB Vietnamese v1 · STS · Vietnamese80.6#10
VN-MTEB Vietnamese v1 · Retrieval · Vietnamese37.6#12
VN-MTEB Vietnamese v1 · Clustering · Vietnamese45.8#14
VN-MTEB Vietnamese v1 · Reranking · Vietnamese64.1#14
VN-MTEB Vietnamese v1 · Pair Classification · Vietnamese83.9#10
MTEB Russian v1.1 · Classification · Russian60.9#42
MTEB Russian v1.1 · Reranking · Russian69.7#16
MTEB Russian v1.1 · Clustering · Russian52.2#57
MTEB Russian v1.1 · Pair Classification · Russian58.4#54
MTEB Russian v1.1 · Multilabel Classification · Russian36#44
RTEB French beta · Mean over tasks · multilingual (2 languages)47#52
RTEB German beta · Mean over tasks · German43.2#81
RTEB Health beta · Mean over tasks · multilingual (4 languages)54.2#44
RTEB Law beta · Mean over tasks · multilingual (4 languages)38.1#88
MTEB multilingual v2 · Instruction Reranking · English-3.1#165
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)20.4#57

FAQ

How much does Multilingual E5 Large cost?›

$0.01 per 1M tokens.

Is Multilingual E5 Large free to use?›

There is no free API; the cheapest paid price is $0.01 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Multilingual E5 Large, and what is it best at?›

#94 of 200 in Embeddings.

Is Multilingual E5 Large better than Gte Multilingual Base?›

On MTEB Portuguese v1 · STS · Portuguese, Multilingual E5 Large scores 79.2 vs 80.4 — lower.

How can I use Multilingual E5 Large?›

Through the intfloat API.

When was Multilingual E5 Large released? Is there a newer version?›

Released 2024-02-08. It is the current version in its line.

Is Multilingual E5 Large open source?›

Yes, the weights are open (MIT licence).

Can I run Multilingual E5 Large locally?›

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

What is Multilingual E5 Large?›

Multilingual E5 Large is an open-weights embeddings & retrieval model from intfloat, released Feb 2024.

What are the best alternatives to Multilingual E5 Large?›

EmbeddingGemma 300M (cheaper), Multilingual E5 Large Instruct (better), E5 Large Matryoshka Sts Pt (open weights).