ColSmol 256M measured on 5 of 195 Multimodal benchmarks5
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
73.4MIEB lite
Jan 2025released
8K tokensMax input
enLanguages
Ranking?
Full ranking →···
39Colmodernvbert26.140ColSmol 256M21.441Siglip Large Patch16 38416.4Strong at
Alternatives?
ModelScoreCostPriceWhat changes
BetterNemotron 3.5 Embed 1B BF16Open×2.7—$0.015Memory (4-bit): 0.2 GB → 0.9 GB
OpenEVIE 8B×3.1——Languages: not verified+1
BetterJina Embeddings v5 Omni NanounstatedOpen×2.1—$0.02Languages: not verified+1BetterNemotron 3.5 Embed 8B BF16Open×2.8—$0.035Memory (4-bit): 0.2 GB → 5 GBBetterJina Embeddings v5 Omni SmallunstatedOpen×2.1—$0.05Languages: not verified+1ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Nemotron 3.5 Embed 1B BF16Open×2.7—$0.015Memory (4-bit): 0.2 GB → 0.9 GBJina Embeddings v5 Omni NanounstatedOpen×2.1—$0.02Languages: not verified+1Nemotron 3.5 Embed 8B BF16Open×2.8—$0.035Memory (4-bit): 0.2 GB → 5 GBJina Embeddings v5 Omni SmallunstatedOpen×2.1—$0.05Languages: not verified+1
Qwen 3 VL Embedding 8BOpen×2.7—$0.09Languages: not verified+1ModelScoreCostPriceWhat changes
EVIE 8BOpen×3.1——Languages: not verified+1EVIE 4.5BOpen×3.1——Languages: not verified+1wWebAI ColVec1.1 8BOpen×3——Languages: not verified+1
pplx Embed v2 Late 9BOpen×3——Max input: 8K tokens → 4K tokens+2vVultronRetrieverPrime Qwen3.5 8BOpen×3——Memory (4-bit): 0.2 GB → 5 GBModelScoreCostPriceWhat changes
vColqwen2.5 v0.2Open×2.4——Memory (4-bit): 0.2 GB → 1.8 GBvColqwen2 v1.0Open×2.1——Memory (4-bit): 0.2 GB → 1.3 GBvColpali v1.3Open×2——Memory (4-bit): 0.2 GB → 1.8 GBvColSmol 500MOpenn/a——Memory (4-bit): 0.2 GB → 0.3 GBvColpali v1.2Openn/a——Memory (4-bit): 0.2 GB → 1.8 GBvColpali v1.1Openn/a——Memory (4-bit): 0.2 GB → 1.8 GBAll 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
Benchmarks
in Embeddings & RetrievalFAQ
How much does ColSmol 256M cost?›
There is no per-unit API price: self-host only.
Is ColSmol 256M free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is ColSmol 256M, and what is it best at?›
#49 of 89 in Multimodal.
Is ColSmol 256M better than Colmodernvbert?›
On ViDoRe v3 · Mean over tasks · multilingual (6 languages), ColSmol 256M scores 21.4 vs 26.1 — lower.
When was ColSmol 256M released? Is there a newer version?›
Released 2025-01-22. It is the current version in its line.
Is ColSmol 256M open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run ColSmol 256M locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is ColSmol 256M?›
ColSmol 256M is an open-weights embeddings & retrieval model from vidore, released Jan 2025.
What are the best alternatives to ColSmol 256M?›
Nemotron 3.5 Embed 1B BF16 (better), EVIE 8B (open weights).