app.py

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load Qwen model
model_name = "Qwen/Qwen-7B"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

def chat(prompt):
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(**inputs, max_length=100)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Example
prompt = "Hello, how are you?"
print(chat(prompt))