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.
#200 of 200 in Embeddings.
On MTEB English v2 · Retrieval · English, Ternary Weight Embedding scores 1.8 vs 2.1 — about the same.
Released 2024-10-23. It is the current version in its line.
Yes, the weights are open.
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Ternary Weight Embedding is an open-weights embeddings & retrieval model from malenia1, released Oct 2024.
EmbeddingGemma 300M (better), Yuan Embedding 2.0 En (open weights).
BetterEmbeddingGemma 300MunstatedOpen×31.2—$0.002Memory (4-bit): 0.1 GB → 0.2 GBIOpenYuan Embedding 2.0 Enunstated×39.7——Memory (4-bit): 0.1 GB → 0.4 GB
BetterBge Base En v1.5unstatedOpen×30.7—$0.005tBetterGte BaseunstatedOpen×29.1—$0.005iBetterE5 Base v2unstatedOpen×27.9—$0.005
Qwen 3 Embedding 8BunstatedOpen×39—$0.01Memory (4-bit): 0.1 GB → 4.5 GB