There is no per-unit API price: price not verified.
#73 of 200 in Embeddings.
On MTEB French v1 · STS · French, GIST Embedding v0 scores 72.1 vs 72.3 — about the same.
Released 2024-01-31. 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.
GIST Embedding v0 is an open-weights embeddings & retrieval model from avsolatorio, released Jan 2024.
Qwen 3 Embedding 8B (better).
BetterQwen 3 Embedding 8BunstatedOpen+20%—$0.01Languages: not verified+1iBetterMultilingual E5 Large InstructunstatedOpen+15%—$0.01Memory (4-bit): 0.1 GB → 0.3 GBiBetterE5 Large v2unstatedOpen+4%—$0.01Memory (4-bit): 0.1 GB → 0.2 GBtBetterGte LargeunstatedOpen+4%—$0.01Memory (4-bit): 0.1 GB → 0.2 GB
Harrier OSS v1 0.6BunstatedOpen+16%——Memory (4-bit): 0.1 GB → 0.4 GBcF2LLM v2 4BunstatedOpen+16%——Memory (4-bit): 0.1 GB → 2.4 GB