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.
#122 of 200 in Embeddings.
On MTEB multilingual v2 · Multilabel Classification · multilingual (25 languages), Bilingual Embedding Large scores 22.4 vs 22.7 — about the same.
Released 2024-06-24. It is the current version in its line.
Yes, the weights are open (Apache-2.0 licence).
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Bilingual Embedding Large is an open-weights embeddings & retrieval model from Lajavaness, released Jun 2024.
EmbeddingGemma 300M (better), Jina Embeddings v5 Text Small (open weights).