app.py

import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "Qwen/Qwen2.5-Coder-14B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, load_in_4bit=True, device_map="auto")

def chat(message, history):
    messages = [{"role": "user", "content": message}]
    text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
    inputs = tokenizer(text, return_tensors="pt").to(model.device)
    with torch.no_grad():
        outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.7)
    response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
    history.append((message, response))
    return history, ""

demo = gr.ChatInterface(chat)
demo.launch()