from transformers import AutoTokenizer, AutoModel

kernel_file = ".\\models\\quantization_kernels.so"
tokenizer = AutoTokenizer.from_pretrained(".\\", trust_remote_code=True)
model = AutoModel.from_pretrained(".\\", trust_remote_code=True).float()

# model = model.quantize(bits=4, kernel_file=kernel_file)

# response, history = model.chat(tokenizer, "你好", history=[])
# print("response:", response)


tokenizer.save_pretrained("models")
model.save_pretrained("models")