There is no per-unit API price: self-host only.
There is no free API. The weights are open, so you can run it yourself.
#67 of 89 in Multimodal.
On MIEB multilingual · Document Understanding · English, Clip Vit Large Patch14 scores 38 vs 38.6 — about the same.
Released 2021-02-26. It is the current version in its line.
Yes, the weights are open.
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Clip Vit Large Patch14 is an open-weights embeddings & retrieval model from OpenAI, released Feb 2021.
Jina Embeddings v5 Omni Nano (better), Argus Colqwen3.5 9B v0 (open weights).
BetterJina Embeddings v5 Omni NanounstatedOpen×2.1—$0.02Memory (4-bit): 0.3 GB → 0.6 GBDOpenArgus Colqwen3.5 9B v0unstated×2.4——Memory (4-bit): 0.3 GB → 5.3 GB
Nemotron Colembed VL 8B v2Open×2.4——Memory (4-bit): 0.3 GB → 5.2 GBDArgus Colqwen3.5 4B v0unstatedOpen×2.4——Memory (4-bit): 0.3 GB → 2.8 GBvVultronRetrieverCore Qwen3.5 4.5BOpen×2.4——Memory (4-bit): 0.3 GB → 2.7 GBvVultronRetrieverPrime Qwen3.5 8BOpen×2.4——Memory (4-bit): 0.3 GB → 5 GB