Gte Qwen2 7B Instruct

Jun 2024
Gte Qwen2 7B Instruct
Compare in
on
self-host onlyprice
58.1MTEB English v2
Jun 2024released
3,584Dimensions
33K tokensMax input

Rankingconfiguration unstated?

Full ranking →
1F2LLM v2 14B73.92F2LLM v2 8B733F2LLM v2 4B72.14Hakim71.65F2LLM v2 1.7B70.4
···
11Jina Embeddings v365.312Gte Qwen2 7B Instruct64.613Multilingual E5 Large64.3

Strong at

A
Embeddings & Retrieval#37 of 200

Alternatives?

ModelScoreCostPriceWhat changes
BetterEmbeddingGemma 300MunstatedOpen+5%—$0.002Max input: 33K tokens → 2K tokenscOpenF2LLM v2 14Bunstated+14%——Memory (4-bit): 4.2 GB → 8.4 GBcBetterF2LLM v2 8BunstatedOpen+13%——Memory (4-bit): 4.2 GB → 4.5 GBcBetterF2LLM v2 4BunstatedOpen+12%——MBetterHakim+11%——Max input: 33K tokens → 1K tokens

Run it yourself

estimate
Fits on✓ 8 GB (phone, laptop) · 8-bit✓ 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 · English58.9#17
BEIR · Mean over task types · English58.9#17
BEIR · Retrieval · English58.9#17
BRIGHT · Mean over tasks · English22.9#2
BRIGHT · Mean over task types · English22.9#2
BRIGHT · Retrieval · English22.9#2
MTEB Mandarin Chinese v1 · Retrieval · Mandarin Chinese75.7#19
MTEB Mandarin Chinese v1 · STS · Mandarin Chinese65.2#7
MTEB Mandarin Chinese v1 · Reranking · Mandarin Chinese69.2#17
MTEB Mandarin Chinese v1 · Clustering · Mandarin Chinese66.1#21
MTEB Mandarin Chinese v1 · Pair Classification · Mandarin Chinese81.2#24
MTEB Persian v2 · Mean over tasks · Persian64.6#12
MTEB Persian v2 · Mean over task types · Persian67.6#13
MTEB Persian v2 · STS · Persian81#3
MTEB Persian v2 · Clustering · Persian63.5#13
MTEB Persian v2 · Classification · Persian63.4#25
MTEB Persian v2 · Retrieval · Persian57.2#9
MTEB Persian v2 · Reranking · Persian69.3#32
MTEB Persian v2 · Bitext Mining · Persian53.3#24
MTEB Persian v2 · Pair Classification · Persian85.3#8
JMTEB v2 · Clustering · Japanese52.2#23
JMTEB v2 · STS · Japanese79.1#51
LongEmbed · Mean over tasks · English47.2#46
LongEmbed · Mean over task types · English47.2#46
LongEmbed · Retrieval · English47.2#46
MTEB German v1 · Clustering · German39.8#12
MTEB German v1 · Classification · German81.4#11
MTEB German v1 · STS · German71.4#18
MTEB German v1 · Pair Classification · German65.3#1
MTEB German v1 · Reranking · German50.2#30
MTEB Europe v1 · Mean over tasks · multilingual (29 languages)63.6#6
MTEB Europe v1 · Mean over task types · multilingual (29 languages)57#5
MTEB Europe v1 · Retrieval · multilingual (25 languages)58.4#12
MTEB Europe v1 · STS · multilingual (8 languages)76.5#4
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)88.3#35
MTEB Europe v1 · Classification · multilingual (23 languages)64.9#7
MTEB Europe v1 · Clustering · multilingual (29 languages)45.9#35
MTEB Europe v1 · Instruction Reranking · English4.9#14
MTEB Europe v1 · Pair Classification · multilingual (11 languages)92.7#2
MTEB Europe v1 · Reranking · multilingual (12 languages)60.7#25
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)20.2#23
MTEB French v1 · Retrieval · multilingual (2 languages)63.9#11
MTEB French v1 · Reranking · French82.7#16
MTEB French v1 · Clustering · French56.2#8
MTEB French v1 · Classification · French81.6#7
MTEB French v1 · Summarization · French31.5#19
MTEB French v1 · Pair Classification · French80.1#1
MTEB Indic v1 · Mean over tasks · multilingual (25 languages)69.4#9
MTEB Indic v1 · Mean over task types · multilingual (25 languages)68.3#9
MTEB Indic v1 · Classification · multilingual (8 languages)68.3#8
MTEB Indic v1 · Retrieval · multilingual (14 languages)85#9
MTEB Indic v1 · Bitext Mining · multilingual (20 languages)73.1#19
MTEB Indic v1 · Reranking · multilingual (2 languages)84.2#48
MTEB Indic v1 · Pair Classification · Hindi81.7#11
MTEB Indic v1 · STS · multilingual (13 languages)41.8#46
MTEB Indic v1 · Clustering · multilingual (21 languages)44.2#32
MTEB Korean v2 · STS · Korean85.8#6
MTEB Korean v2 · Reranking · Korean50.6#44
MTEB Korean v2 · Classification · Korean67#7
MTEB Dutch v1 · Reranking · Dutch89.4#39
MTEB Polish v1 · Pair Classification · Polish88.5#4
MTEB Polish v1 · Classification · Polish74.5#6
MTEB Polish v1 · STS · Polish70.8#3
MTEB Portuguese v1 · Multilabel Classification · Portuguese25.8#11
MTEB Law v1 · Mean over tasks · multilingual (3 languages)54.2#41
MTEB Law v1 · Mean over task types · multilingual (3 languages)54.2#41
MTEB Law v1 · Retrieval · multilingual (3 languages)54.2#41
MTEB Medical v1 · Mean over tasks · multilingual (10 languages)66.6#1
MTEB Medical v1 · Mean over task types · multilingual (10 languages)68.5#2
MTEB Medical v1 · Retrieval · multilingual (10 languages)68.1#2
MTEB Medical v1 · Clustering · English48.3#15
MTEB Medical v1 · Reranking · Mandarin Chinese89.3#15
FollowIR · Mean over tasks · English4.9#14
FollowIR · Mean over task types · English4.9#14
FollowIR · Instruction Reranking · English4.9#14
MTEB Spanish v1 · Pair Classification · Spanish83.8#3
MTEB Spanish v1 · STS · multilingual (2 languages)79.8#7
MTEB Spanish v1 · Clustering · Spanish49.2#13
MTEB Spanish v1 · Reranking · Spanish60.8#28
MTEB Thai v1 · Clustering · Thai53.4#35
MTEB Thai v1 · Classification · Thai49.2#30
MTEB Thai v1 · Pair Classification · Thai84.4#9
MTEB Thai v1 · Reranking · Thai34.8#28
MTEB Scandinavian v1 · Clustering · multilingual (2 languages)49.7#9
MTEB Scandinavian v1 · Classification · multilingual (6 languages)70.8#8
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)71.8#35
VN-MTEB Vietnamese v1 · Classification · Vietnamese69.1#9
VN-MTEB Vietnamese v1 · STS · Vietnamese80.7#9
VN-MTEB Vietnamese v1 · Clustering · Vietnamese52.7#6
VN-MTEB Vietnamese v1 · Reranking · Vietnamese66.9#5
MTEB Russian v1.1 · Classification · Russian73#12
MTEB Russian v1.1 · STS · Russian75#10
MTEB Russian v1.1 · Reranking · Russian64.5#34
MTEB Russian v1.1 · Clustering · Russian66.4#6
MTEB Russian v1.1 · Pair Classification · Russian67.6#7
MTEB Russian v1.1 · Multilabel Classification · Russian44.3#13
RTEB French beta · Mean over tasks · multilingual (2 languages)48.8#47
RTEB German beta · Mean over tasks · German46.1#69
RTEB Law beta · Mean over tasks · multilingual (4 languages)44#69
MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages)62.5#28
MTEB multilingual v2 · Mean over task types · multilingual (1037 languages)55.9#21
MTEB multilingual v2 · Classification · multilingual (119 languages)61.6#35
MTEB multilingual v2 · Clustering · multilingual (208 languages)52.8#23
MTEB multilingual v2 · STS · multilingual (35 languages)74#35
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)85.1#4
MTEB multilingual v2 · Retrieval · multilingual (120 languages)60.1#41
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)73.9#19
MTEB multilingual v2 · Instruction Reranking · English4.9#14
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)25.5#22
MTEB multilingual v2 · Reranking · multilingual (22 languages)65.6#20

FAQ

How much does Gte Qwen2 7B Instruct cost?›

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

Is Gte Qwen2 7B Instruct free to use?›

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

How good is Gte Qwen2 7B Instruct, and what is it best at?›

#37 of 200 in Embeddings. Strongest at: Embeddings & Retrieval.

Is Gte Qwen2 7B Instruct better than Jina Embeddings v3?›

On MTEB Persian v2 · Mean over tasks · Persian, Gte Qwen2 7B Instruct scores 64.6 vs 65.3 — about the same.

When was Gte Qwen2 7B Instruct released? Is there a newer version?›

Released 2024-06-15. It is the current version in its line.

Is Gte Qwen2 7B Instruct open source?›

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

Can I run Gte Qwen2 7B Instruct locally?›

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

What is Gte Qwen2 7B Instruct?›

Gte Qwen2 7B Instruct is an open-weights embeddings & retrieval model from Alibaba (Qwen), released Jun 2024.

What are the best alternatives to Gte Qwen2 7B Instruct?›

EmbeddingGemma 300M (better), F2LLM v2 14B (open weights).