EADST

LLAMA Model Save with INT8 Format

from transformers import BitsAndBytesConfig
from transformers import AutoModelForCausalLM

config = BitsAndBytesConfig(
    load_in_8bit=True,
)
path = "/home/llm/model/path/"
model = AutoModelForCausalLM.from_pretrained(path, device_map="cpu", quantization_config=config)
model.save_pretrained("model_save_folder-8bit")