from mlx_lm import load, generate model, tokenizer = load("mlx-community/SmolLM-1.7B-Instruct-8bit") response = generate(model, tokenizer, prompt="hello", verbose=True) print(response)
