UME R1 7B measured on 13 of 195 Multimodal benchmarks13
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
8.8MVEB text video v1
Nov 2025released
3,584Dimensions
33K tokensMax input
enLanguages
Alternatives?
ModelScoreCostPriceWhat changes
LBetterLCO Embedding Omni 3BunstatedOpen+23%——Languages: not verifiedLBetterLCO Embedding Omni 7BunstatedOpen+15%——Languages: not verified+1VBetterVLM2Vec v2.0Open+9%——HOpenE5 Omni 3Bunstated+6%——Languages: not verifiedModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
LLCO Embedding Omni 3BunstatedOpen+23%——Languages: not verifiedLLCO Embedding Omni 7BunstatedOpen+15%——Languages: not verified+1VVLM2Vec v2.0Open+9%——Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 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-V2lab64.1
FAQ
How much does UME R1 7B cost?›
There is no per-unit API price: self-host only.
Is UME R1 7B free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is UME R1 7B, and what is it best at?›
#16 of 27 in Multimodal.
Is UME R1 7B better than Qwen 3 VL Embedding 2B?›
On MVEB text video v1 · Video Centric QA · English, UME R1 7B scores 26.7 vs 26.7 — about the same.
When was UME R1 7B released? Is there a newer version?›
Released 2025-11-10. It is the current version in its line.
Is UME R1 7B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run UME R1 7B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is UME R1 7B?›
UME R1 7B is an open-weights embeddings & retrieval model from zhibinlan, released Nov 2025.
What are the best alternatives to UME R1 7B?›
LCO Embedding Omni 3B (better).