import llm

# This takes model IDs and aliases defined by plugins:
model = llm.get_embedding_model("sentence-transformers/all-MiniLM-L6-v2")
vector = model.embed("This is text to embed")