Dinov2 Base
Jul 2023
Dinov2 BaseCompare inon
Dinov2 Base measured on 18 of 203 Multimodal benchmarks18
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
19.3MIEB lite
Jul 2023released
768Dimensions
Alternatives?
ModelScoreCostPriceWhat changes
BetterEmbeddingGemma 2unstatedOpen×4—$0.02Memory (4-bit): 0.1 GB → 0.2 GBLOpenLCO Embedding Omni 7Bunstated×4.3——Memory (4-bit): 0.1 GB → 5.4 GB
BetterJina Embeddings v5 Omni NanounstatedOpen×2.8—$0.02Memory (4-bit): 0.1 GB → 0.6 GBBetterJina Embeddings v5 Omni SmallunstatedOpen×3.4—$0.05Memory (4-bit): 0.1 GB → 1 GBBetterJina Clip v1Open+79%—$0.05ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
EmbeddingGemma 2unstatedOpen×4—$0.02Memory (4-bit): 0.1 GB → 0.2 GBJina Embeddings v5 Omni NanounstatedOpen×2.8—$0.02Memory (4-bit): 0.1 GB → 0.6 GBJina Embeddings v5 Omni SmallunstatedOpen×3.4—$0.05Memory (4-bit): 0.1 GB → 1 GBJina Clip v1Open+79%—$0.05VVoyage Multimodal 3×3.7—$0.12ModelScoreCostPriceWhat changes
LLCO Embedding Omni 7BunstatedOpen×4.3——Memory (4-bit): 0.1 GB → 5.4 GBLLCO Embedding Omni 3BunstatedOpen×4.1——Memory (4-bit): 0.1 GB → 2.8 GBHE5 Omni 7BunstatedOpen×4——Memory (4-bit): 0.1 GB → 5.4 GBEmbeddingGemma 2unstatedOpen×4—$0.02Memory (4-bit): 0.1 GB → 0.2 GBHE5 Omni 3BunstatedOpen×3.8——Memory (4-bit): 0.1 GB → 2.8 GBModelScoreCostPriceWhat changes
Dinov2 LargeOpensame——Memory (4-bit): 0.1 GB → 0.2 GBDinov2 GiantOpen-8%——Memory (4-bit): 0.1 GB → 0.7 GBDinov2 SmallOpen-9%——Pe Av BaseOpenn/a——Memory (4-bit): 0.1 GB → 0.6 GBPe Av LargeOpenn/a——Memory (4-bit): 0.1 GB → 1.3 GBPe Av SmallOpenn/a——Memory (4-bit): 0.1 GB → 0.5 GBAll 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 Dinov2 Base cost?›
There is no per-unit API price: self-host only.
Is Dinov2 Base free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Dinov2 Base, and what is it best at?›
#51 of 56 in Multimodal.
Is Dinov2 Base better than Dinov2 Large?›
On MIEB multilingual · Visual semantic similarity (multilingual) · multilingual (13 languages), Dinov2 Base scores 19.3 vs 19.9 — about the same.
When was Dinov2 Base released? Is there a newer version?›
Released 2023-07-18. It is the current version in its line.
Is Dinov2 Base open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Dinov2 Base locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Dinov2 Base?›
Dinov2 Base is an open-weights embeddings & retrieval model from Meta, released Jul 2023.
What are the best alternatives to Dinov2 Base?›
EmbeddingGemma 2 (better), LCO Embedding Omni 7B (open weights).