H

KaLM Embedding Multilingual Mini v1

Aug 2024
HKaLM Embedding Multilingual Mini v1
Compare in
on
price not verifiedprice
54.2MTEB English v2
Aug 2024released
896Dimensions
1K tokensMax input

Rankingconfiguration unstated?

Full ranking →
1F2LLM v2 14B86.12F2LLM v2 8B85.43F2LLM v2 4B85.34F2LLM v2 1.7B855F2LLM v2 0.6B84.1
···
28KaLM Embedding Multilingual Mini Instruct v167.729KaLM Embedding Multilingual Mini v167.630Sentence Croissant Alpha v0.367.1

Strong at

B
Embeddings & Retrieval#68 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

Benchmarks

in Embeddings & Retrieval
BEIR · Mean over tasks · English50.8#96
BEIR · Mean over task types · English50.8#96
BEIR · Retrieval · English50.8#96
BEIR-NL · Mean over tasks · Dutch40.5#10
BEIR-NL · Mean over task types · Dutch40.5#10
BEIR-NL · Retrieval · Dutch40.5#10
MTEB Mandarin Chinese v1 · Retrieval · Mandarin Chinese69.8#48
MTEB Mandarin Chinese v1 · STS · Mandarin Chinese48#49
MTEB Mandarin Chinese v1 · Reranking · Mandarin Chinese64.8#44
MTEB Mandarin Chinese v1 · Clustering · Mandarin Chinese49.9#46
MTEB Mandarin Chinese v1 · Pair Classification · Mandarin Chinese73.2#36
MTEB Persian v2 · Reranking · Persian66.7#40
JMTEB v2 · Clustering · Japanese46.1#51
JMTEB v2 · STS · Japanese80#42
LongEmbed · Mean over tasks · English43.3#56
LongEmbed · Mean over task types · English43.3#56
LongEmbed · Retrieval · English43.3#56
MTEB German v1 · Mean over tasks · German53.4#20
MTEB German v1 · Mean over task types · German52.8#20
MTEB German v1 · Clustering · German34.9#22
MTEB German v1 · Retrieval · German48.5#26
MTEB German v1 · Classification · German65.4#38
MTEB German v1 · STS · German66.6#39
MTEB German v1 · Pair Classification · German52.9#45
MTEB German v1 · Reranking · German48.7#34
MTEB Europe v1 · Retrieval · multilingual (25 languages)50.6#39
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)83#69
MTEB Europe v1 · Classification · multilingual (23 languages)58#30
MTEB Europe v1 · Clustering · multilingual (29 languages)40.4#61
MTEB Europe v1 · Instruction Reranking · English-1.5#105
MTEB Europe v1 · Pair Classification · multilingual (11 languages)87.4#35
MTEB Europe v1 · Reranking · multilingual (12 languages)56.8#62
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)16.9#47
MTEB French v1 · Retrieval · multilingual (2 languages)53#28
MTEB French v1 · Reranking · French80.7#23
MTEB French v1 · Classification · French67.6#29
MTEB French v1 · Summarization · French30.5#53
MTEB French v1 · Pair Classification · French60.6#66
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)58.6#37
MTEB Indic v1 · Mean over task types · multilingual (25 languages)54.3#39
MTEB Indic v1 · Classification · multilingual (8 languages)60.4#38
MTEB Indic v1 · Retrieval · multilingual (14 languages)75.7#37
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)51.5#40
MTEB Indic v1 · Reranking · multilingual (2 languages)82.2#68
MTEB Indic v1 · Pair Classification · Hindi67.5#72
MTEB Indic v1 · STS · multilingual (13 languages)18.6#87
MTEB Indic v1 · Clustering · multilingual (21 languages)24.4#81
MTEB Korean v2 · Reranking · Korean52.7#40
MTEB Korean v2 · Classification · Korean57.3#34
MTEB Dutch v1 · Mean over tasks · Dutch53.5#44
MTEB Dutch v1 · Mean over task types · Dutch60.3#38
MTEB Dutch v1 · Classification · Dutch58.8#46
MTEB Dutch v1 · Clustering · Dutch33.8#39
MTEB Dutch v1 · Retrieval · Dutch55.6#45
MTEB Dutch v1 · STS · Dutch75.2#52
MTEB Dutch v1 · Multilabel Classification · Dutch36.1#32
MTEB Dutch v1 · Pair Classification · Dutch75.8#47
MTEB Dutch v1 · Reranking · Dutch87.1#73
MTEB Polish v1 · Pair Classification · Polish79.9#50
MTEB Polish v1 · Classification · Polish60.8#30
MTEB Portuguese v1 · Multilabel Classification · Portuguese18.7#126
MTEB Law v1 · Mean over tasks · multilingual (3 languages)53.9#43
MTEB Law v1 · Mean over task types · multilingual (3 languages)53.9#43
MTEB Law v1 · Retrieval · multilingual (3 languages)53.9#43
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)57.3#26
MTEB Medical v1 · Mean over task types · multilingual (10 languages)58.7#25
MTEB Medical v1 · Retrieval · multilingual (10 languages)59.8#27
MTEB Medical v1 · Clustering · English33.5#152
MTEB Medical v1 · Reranking · Mandarin Chinese82.7#42
FollowIR · Mean over tasks · English-1.5#105
FollowIR · Mean over task types · English-1.5#105
FollowIR · Instruction Reranking · English-1.5#105
MTEB Spanish v1 · Pair Classification · Spanish67.8#52
MTEB Spanish v1 · Reranking · Spanish59#38
MTEB Thai v1 · Clustering · Thai40.5#70
MTEB Thai v1 · Pair Classification · Thai68.9#61
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)40.8#46
MTEB Scandinavian v1 · Classification · multilingual (6 languages)56.7#57
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)68#61
MTEB Russian v1.1 · Classification · Russian62.1#36
MTEB Russian v1.1 · Reranking · Russian63.2#41
MTEB Russian v1.1 · Clustering · Russian55.7#39
MTEB Russian v1.1 · Pair Classification · Russian57.7#62
MTEB Russian v1.1 · Multilabel Classification · Russian36.1#43
RTEB French beta · Mean over tasks · multilingual (2 languages)40.9#68
RTEB German beta · Mean over tasks · German41.9#82
RTEB Law beta · Mean over tasks · multilingual (4 languages)43.5#73
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)77.7#61
MTEB multilingual v2 · Retrieval · multilingual (120 languages)54.2#70
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)64.8#56
MTEB multilingual v2 · Instruction Reranking · English-1.5#105
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)20.7#56
MTEB multilingual v2 · Reranking · multilingual (22 languages)60.6#61

FAQ

How much does KaLM Embedding Multilingual Mini v1 cost?›

There is no per-unit API price: price not verified.

How good is KaLM Embedding Multilingual Mini v1, and what is it best at?›

#68 of 200 in Embeddings.

Is KaLM Embedding Multilingual Mini v1 better than KaLM Embedding Multilingual Mini Instruct v1?›

On MTEB French v1 · Classification · French, KaLM Embedding Multilingual Mini v1 scores 67.6 vs 67.7 — about the same.

When was KaLM Embedding Multilingual Mini v1 released? Is there a newer version?›

Released 2024-08-27. It is the current version in its line.

Is KaLM Embedding Multilingual Mini v1 open source?›

Yes, the weights are open (MIT licence).

Can I run KaLM Embedding Multilingual Mini v1 locally?›

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

What is KaLM Embedding Multilingual Mini v1?›

KaLM Embedding Multilingual Mini v1 is an open-weights embeddings & retrieval model from HIT-TMG, released Aug 2024.

What are the best alternatives to KaLM Embedding Multilingual Mini v1?›

E5 Mistral 7B Instruct (better), F2LLM v2 8B (open weights).