device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

model_int8 = AutoModelForCausalLM.from_pretrained(model_id,
                                             device_map='auto',
                                             load_in_8bit=True,
                                             )
print(f"Model size: {model_int8.get_memory_footprint():,} bytes")