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.
#82 of 200 in Embeddings.
On MTEB English v2 · Classification · English, Stella En 400M v5 scores 88.2 vs 88.4 — about the same.
Released 2024-07-12. 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.
Stella En 400M v5 is an open-weights embeddings & retrieval model from NovaSearch, released Jul 2024.
Qwen 3 Embedding 8B (better), F2LLM 4B (open weights).
BetterQwen 3 Embedding 8BunstatedOpen+2%—$0.01Memory (4-bit): 0.3 GB → 4.5 GBcOpenF2LLM 4Bunstated+4%——Memory (4-bit): 0.3 GB → 2.4 GBVBetterVoyage 3 M Expunstated+8%——cBetterF2LLM v2 14BunstatedOpen+4%——Memory (4-bit): 0.3 GB → 8.4 GBcBetterF2LLM v2 4BunstatedOpen+4%——Memory (4-bit): 0.3 GB → 2.4 GB