There is no per-unit API price: API price not public.
There is no free API. The weights are open, so you can run it yourself.
#107 of 200 in Embeddings.
On MTEB Law v1 · Mean over task types · multilingual (3 languages), Bekko Embedding v1 A25m scores 57.6 vs 57.9 — about the same.
Released 2026-07-28. 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.
Bekko Embedding v1 A25m is an open-weights embeddings & retrieval model from hotchpotch, released Jul 2026.
Nemotron 3 Embed 1B BF16 (better), Octen Law 8B v1 (open weights).
BetterNemotron 3 Embed 1B BF16Open+19%—$0.015Memory (4-bit): 0.1 GB → 0.7 GBlOpenOcten Law 8B v1+33%——Memory (4-bit): 0.1 GB → 4.5 GBVBetterVoyage 3 Liteunstated+5%—$0.02VBetterVoyage 3unstated+11%—$0.06VBetterVoyage Law 2unstated+13%—$0.12