$0.01 per 1M tokens.
There is no free API; the cheapest paid price is $0.01 per 1M input tokens. The weights are open, so you can run it yourself.
#112 of 200 in Embeddings.
On RTEB Law beta · Mean over tasks · multilingual (4 languages), E5 Large v2 scores 33.9 vs 34 — about the same.
Through the intfloat API.
Released 2024-02-08. 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.
E5 Large v2 is an open-weights embeddings & retrieval model from intfloat, released Feb 2024.
Bge Base En v1.5 (cheaper), Qwen 3 Embedding 8B (better), Octen Embedding 8B (open weights).
CheaperBge Base En v1.5unstatedOpen-4%-50%$0.005
BetterQwen 3 Embedding 8BunstatedOpen+98%same$0.01Languages: not verified+1OOpenOcten Embedding 8B×2.3——Memory (4-bit): 0.2 GB → 4.5 GBtCheaperGte BaseunstatedOpen-5%-50%$0.005iCheaperE5 Base v2unstatedOpen-6%-50%$0.005
Nemotron 3 Embed 1B BF16Open+81%+50%$0.015Memory (4-bit): 0.2 GB → 0.7 GB