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.
#75 of 89 in Multimodal.
On ViDoRe v1&v2 · Mean over tasks · multilingual (4 languages), Clip Vit Base Patch16 scores 21.4 vs 36.4 — lower.
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 Base Patch16 is an open-weights embeddings & retrieval model from OpenAI, released Feb 2021.
Jina Embeddings v4 (better), Argus Colqwen3.5 9B v0 (open weights).
BetterJina Embeddings v4unstatedOpen×3.8—$0.05Memory (4-bit): 0.1 GB → 2.4 GBDOpenArgus Colqwen3.5 9B v0unstated×4——Memory (4-bit): 0.1 GB → 5.3 GBVBetterVoyage Multimodal 3×3.5—$0.12vBetterVultronRetrieverPrime Qwen3.5 8BOpen×4——Memory (4-bit): 0.1 GB → 5 GBOBetterOps Colqwen3 4BOpen×4——Memory (4-bit): 0.1 GB → 2.9 GB
Nemotron Colembed VL 8B v2Open×4——Memory (4-bit): 0.1 GB → 5.2 GBvVultronRetrieverCore Qwen3.5 4.5BOpen×4——Memory (4-bit): 0.1 GB → 2.7 GB