There is no per-unit API price: price not verified.
#187 of 200 in Embeddings.
On MTEB Russian v1.1 · STS · Russian, Rubert Tiny2 scores 63 vs 64.4 — lower.
Released 2021-10-28. It is the current version in its line.
Yes, the weights are open (MIT licence).
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Rubert Tiny2 is an open-weights embeddings & retrieval model from cointegrated, released Oct 2021.
EmbeddingGemma 300M (better), Qwen 3 Embedding 8B (open weights).