Qwen 3 Embedding 8B

Jun 2025
Qwen 3 Embedding 8B
Compare in
on
$0.01per 1M tokens
69.4MTEB English v2
Jun 2025released
3.3 × vLLM H100 (Baseten)Published speed
4,096Dimensions
33K tokensMax input

Rankingconfiguration unstated?

Full ranking →
1F2LLM v2 14B70.52F2LLM v2 8B70.33Querit 4B69.54F2LLM v2 4B69.45Jina Reranker v367.8
···
16Jina Embeddings v5 Text Small65.718Qwen 3 Embedding 8B65.619Gemini Embedding 00165.6

Strong at

S
Embeddings & Retrieval#3 of 200

Alternatives?

ModelScoreCostPriceWhat changes
CheaperEmbeddingGemma 300MunstatedOpen-4%-80%$0.002Max input: 33K tokens → 2K tokenscBetterF2LLM v2 14BunstatedOpen+7%——Memory (4-bit): 4.5 GB → 8.4 GBcBetterF2LLM v2 8BunstatedOpen+7%——cBetterF2LLM v2 4BunstatedOpen+6%——aBetterGiga Embeddings Instruct 10B A1.8B 0826unstatedOpen+3%——Max input: 33K tokens → 8K tokens+1

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 · English62.8#4
BEIR · Mean over task types · English62.8#4
BEIR · Retrieval · English62.8#4
MTEB Mandarin Chinese v1 · Mean over tasks · Mandarin Chinese73.9#7
MTEB Mandarin Chinese v1 · Mean over task types · Mandarin Chinese75#8
MTEB Mandarin Chinese v1 · Retrieval · Mandarin Chinese78.2#9
MTEB Mandarin Chinese v1 · STS · Mandarin Chinese63.5#13
MTEB Mandarin Chinese v1 · Reranking · Mandarin Chinese67#30
MTEB Mandarin Chinese v1 · Clustering · Mandarin Chinese80.1#3
MTEB Mandarin Chinese v1 · Pair Classification · Mandarin Chinese84.2#15
MTEB Mandarin Chinese v1 · Classification · Mandarin Chinese77.1#10
MTEB Europe v1 · Bitext Mining · multilingual (28 languages)93#4
MTEB Europe v1 · Clustering · multilingual (29 languages)50.4#12
MTEB Europe v1 · Instruction Reranking · English10.1#4
MTEB Europe v1 · Reranking · multilingual (12 languages)62.6#11
MTEB Europe v1 · Multilabel Classification · multilingual (23 languages)23.7#8
MTEB French v1 · Pair Classification · French79.2#3
MTEB Indic v1 · Reranking · multilingual (2 languages)86.8#25
MTEB Indic v1 · Pair Classification · Hindi86.6#4
MTEB Indic v1 · STS · multilingual (13 languages)61.6#7
MTEB Indic v1 · Clustering · multilingual (21 languages)55.7#9
MTEB Korean v2 · Mean over tasks · multilingual (2 languages)77.7#1
MTEB Korean v2 · Mean over task types · multilingual (2 languages)75.4#1
MTEB Korean v2 · STS · Korean89.8#2
MTEB Korean v2 · Retrieval · multilingual (2 languages)78.3#3
MTEB Korean v2 · Pair Classification · Korean77.3#1
MTEB Korean v2 · Reranking · Korean67.4#6
MTEB Korean v2 · Classification · Korean69.1#3
MTEB Korean v2 · Clustering · Korean70.9#6
MTEB Dutch v1 · Mean over tasks · Dutch65.6#5
MTEB Dutch v1 · Mean over task types · Dutch70.1#3
MTEB Dutch v1 · Classification · Dutch73#5
MTEB Dutch v1 · Clustering · Dutch49#6
MTEB Dutch v1 · Retrieval · Dutch66.8#5
MTEB Dutch v1 · STS · Dutch87#2
MTEB Dutch v1 · Multilabel Classification · Dutch41.5#5
MTEB Dutch v1 · Pair Classification · Dutch82.4#4
MTEB Dutch v1 · Reranking · Dutch91.2#21
MTEB Portuguese v1 · Multilabel Classification · Portuguese22#40
MTEB Medical v1 · Clustering · English45.1#28
MTEB Medical v1 · Reranking · Mandarin Chinese86.4#31
FollowIR · Mean over tasks · English10.1#4
FollowIR · Mean over task types · English10.1#4
FollowIR · Instruction Reranking · English10.1#4
MTEB Spanish v1 · Pair Classification · Spanish84.4#2
MTEB Thai v1 · Clustering · Thai58.2#16
MTEB Thai v1 · Pair Classification · Thai88.6#2
MTEB Scandinavian v1 · Classification · multilingual (6 languages)74.8#3
MTEB Scandinavian v1 · Bitext Mining · multilingual (3 languages)84.7#8
CoIR · Mean over tasks · multilingual (9 languages)79.9#2
CoIR · Mean over task types · multilingual (9 languages)79.9#2
CoIR · Retrieval · multilingual (9 languages)79.9#2
CoREB v1 · Mean over tasks · multilingual (6 languages)59#10
CoREB v1 · Mean over task types · multilingual (6 languages)59#10
CoREB v1 · Reranking · multilingual (6 languages)54.1#10
CoREB v1 · Retrieval · multilingual (6 languages)63.8#10
MTEB Code v1 · Mean over tasks · multilingual (15 languages)80.7#4
MTEB Code v1 · Mean over task types · multilingual (15 languages)80.7#4
MTEB Code v1 · Retrieval · multilingual (15 languages)80.7#4
MTEB Russian v1.1 · Mean over tasks · Russian71.4#4
MTEB Russian v1.1 · Mean over task types · Russian69.2#4
MTEB Russian v1.1 · Classification · Russian74.3#9
MTEB Russian v1.1 · STS · Russian79.3#1
MTEB Russian v1.1 · Reranking · Russian70.2#10
MTEB Russian v1.1 · Clustering · Russian66.7#5
MTEB Russian v1.1 · Pair Classification · Russian71.3#4
MTEB Russian v1.1 · Retrieval · Russian78.2#4
MTEB Russian v1.1 · Multilabel Classification · Russian44.6#12
RTEB Finnish beta · Mean over tasks · English72.7#23
RTEB French beta · Mean over tasks · multilingual (2 languages)59.5#11
RTEB German beta · Mean over tasks · German66.8#24
RTEB Health beta · Mean over tasks · multilingual (4 languages)67#10
RTEB Law beta · Mean over tasks · multilingual (4 languages)67.2#10
MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages)70.6#3
MTEB multilingual v2 · Mean over task types · multilingual (1037 languages)61.7#3
MTEB multilingual v2 · Classification · multilingual (119 languages)74#4
MTEB multilingual v2 · Clustering · multilingual (208 languages)57.6#6
MTEB multilingual v2 · STS · multilingual (35 languages)81.1#2
MTEB multilingual v2 · Pair Classification · multilingual (24 languages)86.4#1
MTEB multilingual v2 · Retrieval · multilingual (120 languages)70.9#6
MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages)80.9#7
MTEB multilingual v2 · Instruction Reranking · English10.1#4
MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages)28.7#12
MTEB multilingual v2 · Reranking · multilingual (22 languages)65.6#18

FAQ

How much does Qwen 3 Embedding 8B cost?›

$0.01 per 1M tokens.

Is Qwen 3 Embedding 8B free to use?›

There is no free API; the cheapest paid price is $0.01 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Qwen 3 Embedding 8B, and what is it best at?›

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

Is Qwen 3 Embedding 8B better than Jina Embeddings v5 Text Small?›

On MTEB multilingual v2 · Reranking · multilingual (22 languages), Qwen 3 Embedding 8B scores 65.6 vs 65.7 — about the same. It is 80% cheaper.

How can I use Qwen 3 Embedding 8B?›

Through the Alibaba (Qwen) API.

When was Qwen 3 Embedding 8B released? Is there a newer version?›

Released 2025-06-05. It is the current version in its line.

Is Qwen 3 Embedding 8B open source?›

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

Can I run Qwen 3 Embedding 8B locally?›

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

What is Qwen 3 Embedding 8B?›

Qwen 3 Embedding 8B is an open-weights embeddings & retrieval model from Alibaba (Qwen), released Jun 2025.

What are the best alternatives to Qwen 3 Embedding 8B?›

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