m

Sentence Croissant Alpha v0.3

Apr 2024
mSentence Croissant Alpha v0.3
Compare in
on
self-host onlyprice
42.6MTEB English v2
Apr 2024released
2,048Dimensions
1K tokensMax input

Rankingconfiguration unstated?

Full ranking →
1Cohere Embed Multilingual v3.055.62Nemotron 3 Embed 1B BF1655.43Jina Embeddings v5 Text Nano55.14Jina Embeddings v5 Text Small54.95Most Embed De54.7
···
62E5 Large v234.963Sentence Croissant Alpha v0.334.664Snowflake Arctic Embed L32.4

Strong at

C
Embeddings & Retrieval#139 of 200

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
Licencenot verified

History

3 versions
34.53535.536Apr 24Release dateMTEB German · Rerankingv0.2v0.3v0.4
34.53535.536Apr 24Release dateMTEB German · Rerankingv0.2v0.3v0.4

Benchmarks

in Embeddings & Retrieval
MTEB Persian v2 · Reranking · Persian37.9#82
JMTEB v2 · Clustering · Japanese27.6#75
JMTEB v2 · STS · Japanese63.4#89
LongEmbed · Mean over tasks · English36.5#97
LongEmbed · Mean over task types · English36.5#97
LongEmbed · Retrieval · English36.5#97
MTEB German v1 · Clustering · German28.9#52
MTEB German v1 · Classification · German59.3#61
MTEB German v1 · STS · German57.6#54
MTEB German v1 · Pair Classification · German52.7#49
MTEB German v1 · Reranking · German34.6#63
MTEB Europe v1 · Mean over tasks · multilingual (29 languages)49.5#35
MTEB Europe v1 · Mean over task types · multilingual (29 languages)44.5#33
MTEB Europe v1 · Retrieval · multilingual (25 languages)42.1#67
MTEB Europe v1 · STS · multilingual (8 languages)65#51
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)59.7#94
MTEB Europe v1 · Classification · multilingual (23 languages)50.8#82
MTEB Europe v1 · Clustering · multilingual (29 languages)35.8#103
MTEB Europe v1 · Instruction Reranking · English-1.2#93
MTEB Europe v1 · Pair Classification · multilingual (11 languages)83.1#87
MTEB Europe v1 · Reranking · multilingual (12 languages)51.4#95
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)13.6#78
MTEB French v1 · Retrieval · multilingual (2 languages)43.5#47
MTEB French v1 · Reranking · French75.5#40
MTEB French v1 · Classification · French67.1#30
MTEB French v1 · Summarization · French32.5#3
MTEB French v1 · Pair Classification · French65.4#26
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)36.3#63
MTEB Indic v1 · Mean over task types · multilingual (25 languages)25.2#71
MTEB Indic v1 · Classification · multilingual (8 languages)48#61
MTEB Indic v1 · Retrieval · multilingual (14 languages)17.8#84
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)2.7#125
MTEB Indic v1 · Reranking · multilingual (2 languages)46.8#138
MTEB Indic v1 · Pair Classification · Hindi56.9#108
MTEB Indic v1 · STS · multilingual (13 languages)0.6#162
MTEB Indic v1 · Clustering · multilingual (21 languages)4#120
MTEB Korean v2 · STS · Korean45.1#53
MTEB Korean v2 · Reranking · Korean17.1#75
MTEB Korean v2 · Classification · Korean24.8#72
MTEB Dutch v1 · Reranking · Dutch78.9#126
MTEB Polish v1 · Pair Classification · Polish66.7#78
MTEB Polish v1 · Classification · Polish47.9#63
MTEB Polish v1 · STS · Polish57.2#63
MTEB Portuguese v1 · Multilabel Classification · Portuguese14.3#189
MTEB Law v1 · Mean over tasks · multilingual (3 languages)43.5#63
MTEB Law v1 · Mean over task types · multilingual (3 languages)43.5#63
MTEB Law v1 · Retrieval · multilingual (3 languages)43.5#63
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)39.4#67
MTEB Medical v1 · Mean over task types · multilingual (10 languages)33.9#64
MTEB Medical v1 · Retrieval · multilingual (10 languages)42.3#70
MTEB Medical v1 · Clustering · English32.1#175
MTEB Medical v1 · Reranking · Mandarin Chinese27.3#110
FollowIR · Mean over tasks · English-1.2#93
FollowIR · Mean over task types · English-1.2#93
FollowIR · Instruction Reranking · English-1.2#93
MTEB Spanish v1 · Pair Classification · Spanish62.2#80
MTEB Spanish v1 · STS · multilingual (2 languages)72#36
MTEB Spanish v1 · Reranking · Spanish45.1#62
MTEB Thai v1 · Clustering · Thai3.9#134
MTEB Thai v1 · Pair Classification · Thai52.9#112
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)35.7#70
MTEB Scandinavian v1 · Classification · multilingual (6 languages)50.5#74
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)53.1#175
MTEB Russian v1.1 · Classification · Russian52.3#76
MTEB Russian v1.1 · STS · Russian62.4#67
MTEB Russian v1.1 · Reranking · Russian47.3#64
MTEB Russian v1.1 · Clustering · Russian46.5#78
MTEB Russian v1.1 · Pair Classification · Russian53.8#98
MTEB Russian v1.1 · Multilabel Classification · Russian28#86
RTEB French beta · Mean over tasks · multilingual (2 languages)30.1#90
RTEB German beta · Mean over tasks · German29.8#101
RTEB Law beta · Mean over tasks · multilingual (4 languages)35.3#97
MTEB multilingual v2 · Clustering · multilingual (208 languages)38.9#109
MTEB multilingual v2 · STS · multilingual (35 languages)59.7#81
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)73#98
MTEB multilingual v2 · Retrieval · multilingual (120 languages)37.3#113
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)31.9#95
MTEB multilingual v2 · Instruction Reranking · English-1.2#93
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)15.7#125
MTEB multilingual v2 · Reranking · multilingual (22 languages)51.5#81

FAQ

How much does Sentence Croissant Alpha v0.3 cost?›

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

Is Sentence Croissant Alpha v0.3 free to use?›

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

How good is Sentence Croissant Alpha v0.3, and what is it best at?›

#139 of 200 in Embeddings.

Is Sentence Croissant Alpha v0.3 better than E5 Large v2?›

On MTEB German v1 · Reranking · German, Sentence Croissant Alpha v0.3 scores 34.6 vs 34.9 — about the same.

When was Sentence Croissant Alpha v0.3 released? Is there a newer version?›

Released 2024-04-26. It is the current version in its line.

Is Sentence Croissant Alpha v0.3 open source?›

Yes, the weights are open.

Can I run Sentence Croissant Alpha v0.3 locally?›

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

What is Sentence Croissant Alpha v0.3?›

Sentence Croissant Alpha v0.3 is an open-weights embeddings & retrieval model from manu, released Apr 2024.

What are the best alternatives to Sentence Croissant Alpha v0.3?›

Multilingual E5 Large (better), Nemotron 3 Embed 1B BF16 (open weights).