import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained('./', config='tokenizer_config.json', vocab_file='vocab.json')
model = AutoModelForCausalLM.from_pretrained('./', weights='model.safetensors')

def test_model(input_text, expected_output):
    inputs = tokenizer(input_text, return_tensors='pt')
    outputs = model.generate(**inputs)
    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    assert generated_text == expected_output, f"Expected: {expected_output}, but got: {generated_text}"

test_model("Hello, how are you?", "Hello, how are you?")