o

Arabic English Sts Matryoshka

Oct 2024
oArabic English Sts Matryoshka
Compare in
on
self-host onlyprice
15.7MTEB English v2
Oct 2024released
1,024Dimensions
1K tokensMax input
en, arLanguages

Rankingconfiguration unstated?

Full ranking →
1Gte Qwen2 7B Instruct65.32German Semantic STS v263.23Gte Qwen1.5 7B Instruct624F2LLM v2 14B59.85F2LLM v2 8B59.1
···
7Jina Embeddings v5 Text Small58.68Arabic English Sts Matryoshka58.49Gte Qwen2 1.5B Instruct57.9

Strong at

C
Embeddings & Retrieval#186 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
LicenceApache-2.0free for commercial use

Benchmarks

in Embeddings & Retrieval
BEIR · Mean over tasks · English16.3#185
BEIR · Mean over task types · English16.3#185
BEIR · Retrieval · English16.3#185
MTEB Persian v2 · Reranking · Persian53.2#59
JMTEB v2 · Clustering · Japanese45#54
JMTEB v2 · STS · Japanese73.5#70
LongEmbed · Mean over tasks · English19.7#152
LongEmbed · Mean over task types · English19.7#152
LongEmbed · Retrieval · English19.7#152
MTEB German v1 · Clustering · German27#64
MTEB German v1 · Classification · German63.2#51
MTEB German v1 · STS · German68.9#27
MTEB German v1 · Pair Classification · German58.4#8
MTEB German v1 · Reranking · German26.9#77
MTEB Europe v1 · Retrieval · multilingual (25 languages)32.5#107
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)83.5#65
MTEB Europe v1 · Classification · multilingual (23 languages)58.4#28
MTEB Europe v1 · Clustering · multilingual (29 languages)36#100
MTEB Europe v1 · Instruction Reranking · English1.4#27
MTEB Europe v1 · Pair Classification · multilingual (11 languages)91.1#4
MTEB Europe v1 · Reranking · multilingual (12 languages)46.5#133
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)14.4#68
MTEB French v1 · Mean over tasks · multilingual (2 languages)52.2#34
MTEB French v1 · Mean over task types · multilingual (2 languages)53.9#29
MTEB French v1 · STS · French79.2#17
MTEB French v1 · Retrieval · multilingual (2 languages)33.1#89
MTEB French v1 · Reranking · French62.9#96
MTEB French v1 · Clustering · French42.7#40
MTEB French v1 · Classification · French63.7#44
MTEB French v1 · Summarization · French30.1#79
MTEB French v1 · Pair Classification · French65.5#25
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)63.6#21
MTEB Indic v1 · Mean over task types · multilingual (25 languages)62.8#21
MTEB Indic v1 · Classification · multilingual (8 languages)64.2#18
MTEB Indic v1 · Retrieval · multilingual (14 languages)63.7#50
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)62.9#31
MTEB Indic v1 · Reranking · multilingual (2 languages)75.8#88
MTEB Indic v1 · Pair Classification · Hindi87.9#2
MTEB Indic v1 · STS · multilingual (13 languages)50.4#28
MTEB Indic v1 · Clustering · multilingual (21 languages)34.8#47
MTEB Korean v2 · Reranking · Korean29.7#61
MTEB Korean v2 · Classification · Korean53.8#45
MTEB Dutch v1 · Mean over tasks · Dutch45.7#62
MTEB Dutch v1 · Mean over task types · Dutch56.3#60
MTEB Dutch v1 · Classification · Dutch58.4#50
MTEB Dutch v1 · Clustering · Dutch30.1#52
MTEB Dutch v1 · Retrieval · Dutch32.2#93
MTEB Dutch v1 · STS · Dutch79.4#29
MTEB Dutch v1 · Multilabel Classification · Dutch33#60
MTEB Dutch v1 · Pair Classification · Dutch82.2#5
MTEB Dutch v1 · Reranking · Dutch79.1#124
MTEB Polish v1 · Pair Classification · Polish87.6#6
MTEB Polish v1 · Classification · Polish61.7#28
MTEB Polish v1 · STS · Polish68#22
MTEB Portuguese v1 · Multilabel Classification · Portuguese17.8#148
MTEB Law v1 · Mean over tasks · multilingual (3 languages)39#96
MTEB Law v1 · Mean over task types · multilingual (3 languages)39#96
MTEB Law v1 · Retrieval · multilingual (3 languages)39#96
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)26.2#108
MTEB Medical v1 · Mean over task types · multilingual (10 languages)31.4#84
MTEB Medical v1 · Retrieval · multilingual (10 languages)23.7#116
MTEB Medical v1 · Clustering · English30.1#196
MTEB Medical v1 · Reranking · Mandarin Chinese40.3#106
FollowIR · Mean over tasks · English1.4#27
FollowIR · Mean over task types · English1.4#27
FollowIR · Instruction Reranking · English1.4#27
MTEB Spanish v1 · Pair Classification · Spanish77.3#10
MTEB Spanish v1 · Clustering · Spanish43.1#53
MTEB Spanish v1 · Reranking · Spanish32.9#91
MTEB Thai v1 · Clustering · Thai40.5#71
MTEB Thai v1 · Pair Classification · Thai88.5#3
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)37.2#61
MTEB Scandinavian v1 · Classification · multilingual (6 languages)58.6#45
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)61.4#129
MTEB Russian v1.1 · Classification · Russian58.3#54
MTEB Russian v1.1 · STS · Russian71.7#39
MTEB Russian v1.1 · Reranking · Russian40.8#73
MTEB Russian v1.1 · Clustering · Russian52.2#56
MTEB Russian v1.1 · Pair Classification · Russian68.3#6
MTEB Russian v1.1 · Multilabel Classification · Russian34.1#50
RTEB French beta · Mean over tasks · multilingual (2 languages)15.5#146
RTEB German beta · Mean over tasks · German28.7#102
RTEB Law beta · Mean over tasks · multilingual (4 languages)31.1#136
MTEB multilingual v2 · Classification · multilingual (119 languages)56.7#60
MTEB multilingual v2 · Clustering · multilingual (208 languages)40.2#96
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)83.4#11
MTEB multilingual v2 · Retrieval · multilingual (120 languages)31.7#141
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)60.9#65
MTEB multilingual v2 · Instruction Reranking · English1.4#27
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)19.3#66
MTEB multilingual v2 · Reranking · multilingual (22 languages)48.3#95

FAQ

How much does Arabic English Sts Matryoshka cost?›

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

Is Arabic English Sts Matryoshka free to use?›

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

How good is Arabic English Sts Matryoshka, and what is it best at?›

#186 of 200 in Embeddings.

Is Arabic English Sts Matryoshka better than Jina Embeddings v5 Text Small?›

On MTEB German v1 · Pair Classification · German, Arabic English Sts Matryoshka scores 58.4 vs 58.6 — about the same.

When was Arabic English Sts Matryoshka released? Is there a newer version?›

Released 2024-10-13. It is the current version in its line.

Is Arabic English Sts Matryoshka open source?›

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

Can I run Arabic English Sts Matryoshka locally?›

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

What is Arabic English Sts Matryoshka?›

Arabic English Sts Matryoshka is an open-weights embeddings & retrieval model from omarelshehy, released Oct 2024.

What are the best alternatives to Arabic English Sts Matryoshka?›

Gte Qwen2 7B Instruct (better).