$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.
#163 of 220 in Embeddings.
On MTEB Medical v1 · Clustering · English, Bge M3 scores 32.2 vs 32.3 — about the same.
Through the BAAI API.
Released 2024-06-28. 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.
Bge M3 is an open-weights embeddings & retrieval model from BAAI, released Jun 2024.
EmbeddingGemma 300M (cheaper), F2LLM 4B (open weights).
CheaperEmbeddingGemma 300MunstatedOpen+34%-80%$0.002Max input: 8K tokens → 2K tokenscOpenF2LLM 4Bunstated×2.2——Memory (4-bit): 0.3 GB → 2.4 GBsCheaperAll Mpnet Base v2unstatedOpen+20%-50%$0.005Max input: 8K tokens → 0K tokenssCheaperAll MiniLM L12 v2unstatedOpen+14%-50%$0.005Max input: 8K tokens → 0K tokenssCheaperAll MiniLM L6 v2unstatedOpen+14%-50%$0.005Max input: 8K tokens → 0K tokens