m

Sentence Croissant Alpha v0.4

Apr 2024
mSentence Croissant Alpha v0.4
Compare in
on
self-host onlyprice
43.2MTEB English v2
Apr 2024released
2,048Dimensions
1K tokensMax input
en, frLanguages

Rankingconfiguration unstated?

Full ranking →
1Gemini Embedding 2 Preview91.72Harrier OSS v1 27B863KaLM Embedding Gemma3 12B 251183.84Harrier OSS v1 0.6B82.85Llama Embed Nemotron 8B81.7
···
99E5 Large30.6100Sentence Croissant Alpha v0.430.3101Sgpt Bloom 7b1 Msmarco29.9

Strong at

C
Embeddings & Retrieval#135 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
LicenceMITfree for commercial use

History

3 versions
3030.53131.53232.5Apr 24Release dateMTEB multilingual · Bitext Miningv0.2v0.3v0.4
3030.53131.53232.5Apr 24Release dateMTEB multilingual · Bitext Miningv0.2v0.3v0.4

Benchmarks

in Embeddings & Retrieval
MTEB Persian v2 · Reranking · Persian39.6#78
JMTEB v2 · Clustering · Japanese27.7#80
JMTEB v2 · STS · Japanese62.3#105
LongEmbed · Mean over tasks · English37.2#92
LongEmbed · Mean over task types · English37.2#92
LongEmbed · Retrieval · English37.2#92
MTEB German v1 · Mean over tasks · German43.6#46
MTEB German v1 · Mean over task types · German43.3#44
MTEB German v1 · Classification · German58.6#66
MTEB German v1 · Pair Classification · German53.4#45
MTEB German v1 · STS · German57#62
MTEB German v1 · Clustering · German29.2#53
MTEB German v1 · Retrieval · German25.6#95
MTEB German v1 · Reranking · German35.8#59
MTEB Europe v1 · Mean over tasks · multilingual (29 languages)49.7#35
MTEB Europe v1 · Mean over task types · multilingual (29 languages)44.7#33
MTEB Europe v1 · Classification · multilingual (23 languages)50.9#79
MTEB Europe v1 · Instruction Reranking · English-0.1#61
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)14.9#62
MTEB Europe v1 · STS · multilingual (8 languages)63.1#74
MTEB Europe v1 · Pair Classification · multilingual (11 languages)82.8#106
MTEB Europe v1 · Clustering · multilingual (29 languages)36.7#89
MTEB Europe v1 · Retrieval · multilingual (25 languages)43.8#69
MTEB Europe v1 · Reranking · multilingual (12 languages)51.7#89
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)58.5#97
MTEB French v1 · Retrieval · multilingual (2 languages)44.3#50
MTEB French v1 · Classification · French67.1#34
MTEB French v1 · Reranking · French76.4#46
MTEB French v1 · Pair Classification · French66.6#21
MTEB French v1 · Summarization · French30.2#75
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)36.2#69
MTEB Indic v1 · Mean over task types · multilingual (25 languages)25.9#69
MTEB Indic v1 · Classification · multilingual (8 languages)47.3#70
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)2.7#129
MTEB Indic v1 · STS · multilingual (13 languages)-0.8#171
MTEB Indic v1 · Reranking · multilingual (2 languages)49.7#126
MTEB Indic v1 · Retrieval · multilingual (14 languages)20.6#85
MTEB Indic v1 · Pair Classification · Hindi57.6#105
MTEB Indic v1 · Clustering · multilingual (21 languages)4.2#114
MTEB Korean v2 · STS · Korean44.6#62
MTEB Korean v2 · Classification · Korean25.3#77
MTEB Korean v2 · Reranking · Korean17.9#76
MTEB Dutch v1 · Reranking · Dutch78.3#132
MTEB Polish v1 · Mean over tasks · Polish51.8#40
MTEB Polish v1 · Mean over task types · Polish51.8#40
MTEB Polish v1 · Classification · Polish48.4#67
MTEB Polish v1 · Clustering · Polish34.7#44
MTEB Polish v1 · STS · Polish57.2#70
MTEB Polish v1 · Pair Classification · Polish66.7#87
MTEB Portuguese v1 · Multilabel Classification · Portuguese13.7#195
MTEB Law v1 · Mean over tasks · multilingual (3 languages)43.9#66
MTEB Law v1 · Mean over task types · multilingual (3 languages)43.9#66
MTEB Law v1 · Retrieval · multilingual (3 languages)43.9#66
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)39.1#72
MTEB Medical v1 · Mean over task types · multilingual (10 languages)33.8#70
MTEB Medical v1 · Retrieval · multilingual (10 languages)41.8#77
MTEB Medical v1 · Clustering · English33.1#164
MTEB Medical v1 · Reranking · Mandarin Chinese26.4#116
FollowIR · Mean over tasks · English-0.1#61
FollowIR · Mean over task types · English-0.1#61
FollowIR · Instruction Reranking · English-0.1#61
MTEB Spanish v1 · STS · multilingual (2 languages)71.8#43
MTEB Spanish v1 · Pair Classification · Spanish62.4#75
MTEB Spanish v1 · Reranking · Spanish44.4#63
MTEB Thai v1 · Pair Classification · Thai52.9#111
MTEB Thai v1 · Clustering · Thai5#116
MTEB Scandinavian v1 · Mean over tasks · multilingual (6 languages)45#70
MTEB Scandinavian v1 · Mean over task types · multilingual (6 languages)44.3#88
MTEB Scandinavian v1 · Classification · multilingual (6 languages)49.7#84
MTEB Scandinavian v1 · Retrieval · multilingual (3 languages)43.1#77
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)35.3#82
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)49.1#190
MTEB Russian v1.1 · Classification · Russian52.2#81
MTEB Russian v1.1 · Clustering · Russian46.6#82
MTEB Russian v1.1 · Reranking · Russian47.7#63
MTEB Russian v1.1 · Multilabel Classification · Russian28.3#89
MTEB Russian v1.1 · STS · Russian62.2#70
MTEB Russian v1.1 · Pair Classification · Russian52.3#116
RTEB French beta · Mean over tasks · multilingual (2 languages)29.7#93
RTEB German beta · Mean over tasks · German31#105
RTEB Law beta · Mean over tasks · multilingual (4 languages)35.4#103
MTEB multilingual v2 · Clustering · multilingual (208 languages)39.8#104
MTEB multilingual v2 · STS · multilingual (35 languages)57.7#85
MTEB multilingual v2 · Retrieval · multilingual (120 languages)38.3#110
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)16.1#116
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)30.3#100
MTEB multilingual v2 · Instruction Reranking · English-0.1#61
MTEB multilingual v2 · Reranking · multilingual (22 languages)51.4#83
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)72.5#107

FAQ

How much does Sentence Croissant Alpha v0.4 cost?›

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

Is Sentence Croissant Alpha v0.4 free to use?›

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

How good is Sentence Croissant Alpha v0.4, and what is it best at?›

#135 of 200 in Embeddings.

Is Sentence Croissant Alpha v0.4 better than E5 Large?›

On MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages), Sentence Croissant Alpha v0.4 scores 30.3 vs 30.6 — about the same.

When was Sentence Croissant Alpha v0.4 released? Is there a newer version?›

Released 2024-04-27. It is the current version in its line.

Is Sentence Croissant Alpha v0.4 open source?›

Yes, the weights are open (MIT licence).

Can I run Sentence Croissant Alpha v0.4 locally?›

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

What is Sentence Croissant Alpha v0.4?›

Sentence Croissant Alpha v0.4 is an open-weights embeddings & retrieval model from manu, released Apr 2024.

What are the best alternatives to Sentence Croissant Alpha v0.4?›

EmbeddingGemma 300M (better), KaLM Embedding Gemma3 12B 2511 (open weights).