Views
No views yet
llama-cpp-python1from llama_cpp import Llama
2import numpy as np
3
4# Load the model in embedding mode
5llm = Llama(model_path="zembed-1-Q4_K_M.gguf", embedding=True)
6
7# Generate an embedding
8text = "How do I optimize a local LLM to run smoothly?"
9embedding = llm.embed(text)