Colqwen2 v1.0 measured on 5 of 203 Multimodal benchmarks5
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
85.7MIEB lite
Nov 2025released
33K tokensMax input
enLanguages
Ranking?
Full ranking →···
33Jina Embeddings v5 Omni Nano45.634Colqwen2 v1.044.735Colpali v1.343.1Strong at
Alternatives?
ModelScoreCostPriceWhat changes
BetterNemotron 3.5 Embed 1B BF16Open+29%—$0.015
OpenEVIE 8B+48%——Languages: not verified+1BetterNemotron 3.5 Embed 8B BF16Open+35%—$0.035Memory (4-bit): 1.3 GB → 5 GB
BetterQwen 3 VL Embedding 8BOpen+29%—$0.09Languages: not verified+1BetterEVIE 4.5BOpen+47%——Languages: not verified+1ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Nemotron 3.5 Embed 1B BF16Open+29%—$0.015Nemotron 3.5 Embed 8B BF16Open+35%—$0.035Memory (4-bit): 1.3 GB → 5 GBQwen 3 VL Embedding 8BOpen+29%—$0.09Languages: not verified+1EVIE 8BOpen+48%——Languages: not verified+1EVIE 4.5BOpen+47%——Languages: not verified+1ModelScoreCostPriceWhat changes
EVIE 8BOpen+48%——Languages: not verified+1EVIE 4.5BOpen+47%——Languages: not verified+1wWebAI ColVec1.1 8BOpen+45%——Languages: not verified+1
pplx Embed v2 Late 9BOpen+45%——Max input: 33K tokens → 4K tokens+2vVultronRetrieverPrime Qwen3.5 8BOpen+44%——Memory (4-bit): 1.3 GB → 5 GBModelScoreCostPriceWhat changes
vColqwen2.5 v0.2Open+16%——Memory (4-bit): 1.3 GB → 1.8 GBvColpali v1.3Open-4%——Max input: 33K tokens → 16K tokens+1vColSmol 256MOpen-52%——Max input: 33K tokens → 8K tokensvColSmol 500MOpenn/a——Max input: 33K tokens → 8K tokensvColpali v1.2Openn/a——Max input: 33K tokens → 16K tokens+1vColpali v1.1Openn/a——Max input: 33K tokens → 16K tokens+1All vidore 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
History
2 versionsBenchmarks
in Embeddings & RetrievalFAQ
How much does Colqwen2 v1.0 cost?›
There is no per-unit API price: self-host only.
Is Colqwen2 v1.0 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Colqwen2 v1.0, and what is it best at?›
#28 of 89 in Multimodal.
Is Colqwen2 v1.0 better than Jina Embeddings v5 Omni Nano?›
On ViDoRe v3 · Mean over tasks · multilingual (6 languages), Colqwen2 v1.0 scores 44.7 vs 45.6 — about the same.
When was Colqwen2 v1.0 released? Is there a newer version?›
Released 2025-11-03. It is the current version in its line.
Is Colqwen2 v1.0 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Colqwen2 v1.0 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Colqwen2 v1.0?›
Colqwen2 v1.0 is an open-weights embeddings & retrieval model from vidore, released Nov 2025.
What are the best alternatives to Colqwen2 v1.0?›
Nemotron 3.5 Embed 1B BF16 (better), EVIE 8B (open weights).