Pe Av Small
Dec 2025
Pe Av SmallCompare inon
Pe Av Small measured on 20 of 195 Multimodal benchmarks20
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
9MVEB text video v1
Dec 2025released
1,024Dimensions
1K tokensMax input
Alternatives?
ModelScoreCostPriceWhat changes
BetterJina Embeddings v5 Omni SmallunstatedOpen+6%—$0.05Memory (4-bit): 0.5 GB → 1 GBLOpenLCO Embedding Omni 3Bunstated+7%——Memory (4-bit): 0.5 GB → 2.8 GBLBetterLCO Embedding Omni 7BunstatedOpen+6%——Memory (4-bit): 0.5 GB → 5.4 GBBBetterBidirLM Omni 2.5B EmbeddingunstatedOpen+5%——Memory (4-bit): 0.5 GB → 1.5 GB
BetterOmni Embed Nemotron 3BOpen+3%——Memory (4-bit): 0.5 GB → 2.8 GBModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Jina Embeddings v5 Omni SmallunstatedOpen+6%—$0.05Memory (4-bit): 0.5 GB → 1 GBLLCO Embedding Omni 3BunstatedOpen+7%——Memory (4-bit): 0.5 GB → 2.8 GBLLCO Embedding Omni 7BunstatedOpen+6%——Memory (4-bit): 0.5 GB → 5.4 GBBBidirLM Omni 2.5B EmbeddingunstatedOpen+5%——Memory (4-bit): 0.5 GB → 1.5 GBOmni Embed Nemotron 3BOpen+3%——Memory (4-bit): 0.5 GB → 2.8 GBModelScoreCostPriceWhat changes
LLCO Embedding Omni 3BunstatedOpen+7%——Memory (4-bit): 0.5 GB → 2.8 GBJina Embeddings v5 Omni SmallunstatedOpen+6%—$0.05Memory (4-bit): 0.5 GB → 1 GBLLCO Embedding Omni 7BunstatedOpen+6%——Memory (4-bit): 0.5 GB → 5.4 GBBBidirLM Omni 2.5B EmbeddingunstatedOpen+5%——Memory (4-bit): 0.5 GB → 1.5 GBOmni Embed Nemotron 3BOpen+3%——Memory (4-bit): 0.5 GB → 2.8 GBModelScoreCostPriceWhat changes
Pe Av BaseOpensame——Memory (4-bit): 0.5 GB → 0.6 GBPe Av LargeOpensame——Memory (4-bit): 0.5 GB → 1.3 GBDinov2 BaseOpenn/a——Max input: not verifiedDinov2 GiantOpenn/a——Max input: not verified+1Dinov2 LargeOpenn/a——Max input: not verifiedDinov2 SmallOpenn/a——Max input: not verifiedAll Meta 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 Pe Av Small cost?›
There is no per-unit API price: self-host only.
Is Pe Av Small free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Pe Av Small, and what is it best at?›
#15 of 27 in Multimodal.
Is Pe Av Small better than Ebind Full?›
On MVEB v1 · Video Pair Classification · English, Pe Av Small scores 75.3 vs 75.4 — about the same.
When was Pe Av Small released? Is there a newer version?›
Released 2025-12-22. It is the current version in its line.
Is Pe Av Small open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Pe Av Small locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Pe Av Small?›
Pe Av Small is an open-weights embeddings & retrieval model from Meta, released Dec 2025.
What are the best alternatives to Pe Av Small?›
Jina Embeddings v5 Omni Small (better), LCO Embedding Omni 3B (open weights).