| from llama_cpp import Llama | |
| import os | |
| model_path = "llama-3.2-1B-it-Ecommerce-ChatBot-merged-F16.gguf" | |
| n_threads = os.cpu_count() | |
| llm = Llama( | |
| model_path=model_path, | |
| n_ctx=512, | |
| n_batch=64, | |
| n_threads=n_threads, | |
| n_gpu_layers=-1, | |
| chat_format="llama-3" | |
| ) | |