Qwen 3 VL Embedding 2B
Jan 2026
Qwen 3 VL Embedding 2BCompare inon
Qwen 3 VL Embedding 2B measured on 15 of 203 Multimodal benchmarks15
BenchmarkByScoreRank
Retrieval1
Other14Leads on 2
No benchmark matches
price not verifiedprice
52.6ViDoRe v3
Jan 2026released
2,048 (or 64)Dimensions
33K tokensMax input
Alternatives?
ModelScoreCostPriceWhat changes
BetterNemotron 3.5 Embed 1B BF16Open+10%—$0.015
OpenEVIE 8B+26%——Memory (4-bit): 1.3 GB → 5 GBBetterNemotron 3.5 Embed 8B BF16Open+14%—$0.035Memory (4-bit): 1.3 GB → 5 GBBetterQwen 3 VL Embedding 8BOpen+10%—$0.09Memory (4-bit): 1.3 GB → 4.9 GBBetterEVIE 4.5BOpen+25%——Memory (4-bit): 1.3 GB → 2.7 GBModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Nemotron 3.5 Embed 1B BF16Open+10%—$0.015Nemotron 3.5 Embed 8B BF16Open+14%—$0.035Memory (4-bit): 1.3 GB → 5 GBQwen 3 VL Embedding 8BOpen+10%—$0.09Memory (4-bit): 1.3 GB → 4.9 GBEVIE 8BOpen+26%——Memory (4-bit): 1.3 GB → 5 GBEVIE 4.5BOpen+25%——Memory (4-bit): 1.3 GB → 2.7 GBModelScoreCostPriceWhat changes
EVIE 8BOpen+26%——Memory (4-bit): 1.3 GB → 5 GBEVIE 4.5BOpen+25%——Memory (4-bit): 1.3 GB → 2.7 GBwWebAI ColVec1.1 8BOpen+23%——Memory (4-bit): 1.3 GB → 5 GB
pplx Embed v2 Late 9BOpen+23%——Max input: 33K tokens → 4K tokens+1vVultronRetrieverPrime Qwen3.5 8BOpen+22%——Memory (4-bit): 1.3 GB → 5 GBModelScoreCostPriceWhat changes
Qwen 3 VL Embedding 8BOpen+10%—$0.09Memory (4-bit): 1.3 GB → 4.9 GBTongyi Embedding Vision Flashn/a—$0.09Max input: not verifiedTongyi Embedding Vision Plusn/a—$0.09Max input: not verifiedGVE 7BOpenn/a——Memory (4-bit): 1.3 GB → 4.9 GBQwen 3 Omni 30B A3B ThinkingOpenn/a——Memory (4-bit): 1.3 GB → 19 GBQwen 2.5 Omni 3BOpenn/a——Memory (4-bit): 1.3 GB → 3.3 GBQwen 2.5 Omni 7BOpenn/a——Memory (4-bit): 1.3 GB → 6.4 GBGme Qwen2 VL 2B InstructunstatedOpenn/a——All Alibaba (Qwen) models →Run it yourself
estimateFits 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 & RetrievalMMEB-V2lab73.3
FAQ
How much does Qwen 3 VL Embedding 2B cost?›
There is no per-unit API price: price not verified.
How good is Qwen 3 VL Embedding 2B, and what is it best at?›
#28 of 50 in Multimodal.
Is Qwen 3 VL Embedding 2B better than Llama Nemoretriever Colembed 1B v1?›
On ViDoRe v3 · Mean over tasks · multilingual (6 languages), Qwen 3 VL Embedding 2B scores 52.6 vs 55.6 — lower.
When was Qwen 3 VL Embedding 2B released? Is there a newer version?›
Released 2026-01-08. It is the current version in its line.
Is Qwen 3 VL Embedding 2B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Qwen 3 VL Embedding 2B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Qwen 3 VL Embedding 2B?›
Qwen 3 VL Embedding 2B is an open-weights embeddings & retrieval model from Alibaba (Qwen), released Jan 2026.
What are the best alternatives to Qwen 3 VL Embedding 2B?›
Nemotron 3.5 Embed 1B BF16 (better), EVIE 8B (open weights).