import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("./")
model = AutoModelForCausalLM.from_pretrained("./", device_map="auto", torch_dtype=torch.float16)

messages = [
    {"role": "user", "content": "9.9 和 9.11 哪个大?"},
]
tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
print(tokenizer.decode(tokenized_chat[0]))

outputs = model.generate(tokenized_chat, max_new_tokens=4096) 
print(tokenizer.decode(outputs[0]))