Update app.py

app.py

@@ -1,63 +1,5 @@
-import torch
 import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
-import os
 
-MODEL_PATH = "zai-org/GLM-4.7-Flash"
+demo = gr.load("Tongyi-MAI/Z-Image-Turbo", src="models")
 
-# 1. Load the model and tokenizer globally for efficiency
-tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
-model = AutoModelForCausalLM.from_pretrained(
-    pretrained_model_name_or_path=MODEL_PATH,
-    torch_dtype=torch.bfloat16,
-    device_map="auto",
-    trust_remote_code=True # Required for many GLM models
-)
-
-def chat_with_glm(message, history):
-    # 2. Format history for the chat template
-    # Gradio history is a list of [user_msg, assistant_msg]
-    messages = []
-    for h in history:
-        messages.append({"role": "user", "content": h[0]})
-        messages.append({"role": "assistant", "content": h[1]})
-    
-    # Add current message
-    messages.append({"role": "user", "content": message})
-
-    # 3. Apply the ChatML-style template
-    inputs = tokenizer.apply_chat_template(
-        messages, 
-        tokenize=True, 
-        add_generation_prompt=True, 
-        return_dict=True, 
-        return_tensors="pt"
-    ).to(model.device)
-
-    # 4. Generate the response
-    generated_ids = model.generate(
-        **inputs, 
-        max_new_tokens=512, 
-        do_sample=True, 
-        top_p=0.9, 
-        temperature=0.7
-    )
-    
-    # 5. Decode and remove the prompt tokens
-    response = tokenizer.decode(
-        generated_ids[0][inputs.input_ids.shape[1]:], 
-        skip_special_tokens=True
-    )
-    
-    return response
-
-# 6. Launch the interface
-demo = gr.ChatInterface(
-    fn=chat_with_glm,
-    title="GLM-4.7-Flash Chatbot",
-    description="Ask anything to the 30B-A3B MoE model from Z.ai.",
-    examples=["Hello!", "Explain Mixture of Experts models.", "Write a Python script for a simple timer."]
-)
-
-if __name__ == "__main__":
-    demo.launch()
+demo.launch(css = """footer {visibility: hidden;}""")