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 Mandarin Chinese v1 · STS · Mandarin Chinese, Ternary Weight Embedding scores 62.7 vs 63.2 — 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.
QZhou Embedding Zh (better).
BetterYoutu EmbeddingOpen+10%——Memory (4-bit): 0.1 GB → 1.6 GB
BetterSeed1.5 Embeddingunstated+6%——TBetterConan Embedding v2unstated+4%——
Gte Qwen2 7B InstructunstatedOpen+4%——Memory (4-bit): 0.1 GB → 4.2 GBlXiaobu Embedding v2Open+2%——Memory (4-bit): 0.1 GB → 0.2 GB