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.
#162 of 200 in Embeddings.
On MTEB French v1 · Pair Classification · French, LLM2Vec Mistral 7B Instruct v2 Mntp Unsup Simcse scores 64.6 vs 64.7 — about the same.
Released 2024-04-09. 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.
LLM2Vec Mistral 7B Instruct v2 Mntp Unsup Simcse is an open-weights embeddings & retrieval model from McGill-NLP, released Apr 2024.
Qwen 3 Embedding 8B (better), Gte Qwen2 7B Instruct (open weights).