Update app.py

app.py

@@ -1,167 +1,5 @@
 import gradio as gr
-import transformers
-import spaces
+import ai_gradio 
 
-title = """# 🙋🏻‍♂️Welcome to 🌟Tonic's 🤳🏻Phi-4 Demo"""
-
-description = """
-    This demo uses Microsoft's Phi-4 model for text generation.
-    - System Prompt: Sets the context/role for the AI
-    - User Prompt: Your specific question or request
-    - Max Tokens: Maximum length of the generated response
-    - Temperature: Controls randomness (higher = more creative, lower = more focused)
-"""
-
-
-join_us = """
-## Join us:
-🌟TeamTonic🌟 is always making cool demos! Join our active builder's 🛠️community 👻 
-[![Join us on Discord](https://img.shields.io/discord/1109943800132010065?label=Discord&logo=discord&style=flat-square)](https://discord.gg/qdfnvSPcqP) 
-On 🤗Huggingface: [MultiTransformer](https://huggingface.co/MultiTransformer) 
-On 🌐Github: [Tonic-AI](https://github.com/tonic-ai) & contribute to🌟 [Dark Thoughts](https://github.com/MultiTonic/thinking-dataset)
-🤗Big thanks to Yuvi Sharma and all the folks at huggingface for the community grant 🤗
-"""
-
-@spaces.GPU(duration=180)
-def generate_response(system_prompt, user_prompt, max_tokens, temperature):
-    pipeline = transformers.pipeline(
-        "text-generation",
-        model="microsoft/phi-4",
-        model_kwargs={"torch_dtype": "auto"},
-        device_map="auto",
-    )
-    
-    messages = [
-        {"role": "system", "content": system_prompt},
-        {"role": "user", "content": user_prompt},
-    ]
-    
-    outputs = pipeline(
-        messages, 
-        max_new_tokens=max_tokens,
-        temperature=temperature,
-        do_sample=True
-    )
-    
-    # Parse the response to get only the assistant's message
-    try:
-        response_text = outputs[0]["generated_text"]
-        # Find the assistant's content
-        assistant_start = response_text.find("'assistant', 'content': '") + len("'assistant', 'content': '")
-        assistant_end = response_text.rfind("'}")
-        if assistant_start > -1 and assistant_end > -1:
-            return response_text[assistant_start:assistant_end]
-        else:
-            return "Could not parse response properly"
-    except Exception as e:
-        return f"Error parsing response: {str(e)}"
-
-# Example configurations
-examples = [
-    [
-        "You are a medieval knight and must provide explanations to modern people.",
-        "How should I explain the Internet?",
-        128,
-        0.7
-    ],
-    [
-        "You are a wise wizard from ancient times.",
-        "What would you call a smartphone?",
-        256,
-        0.8
-    ],
-    [
-        "You are a time-traveling merchant from the year 1400.",
-        "How would you describe modern cars?",
-        200,
-        0.6
-    ],
-    [
-        "You are a medieval monk who specializes in manuscripts.",
-        "What do you think about e-books?",
-        150,
-        0.7
-    ],
-    [
-        "You are a castle guard from the Middle Ages.",
-        "What do you think about modern security systems?",
-        180,
-        0.9
-    ]
-]
-
-# Create the Gradio interface
-with gr.Blocks() as demo:
-    gr.Markdown(title)
-    gr.Markdown(description)
-    gr.Markdown(join_us)
-    
-    with gr.Row():
-        with gr.Column():
-            system_prompt = gr.Textbox(
-                label="🤖Instructions",
-                placeholder="Enter system prompt...",
-                value="You are a medieval knight and must provide explanations to modern people."
-            )
-            user_prompt = gr.Textbox(
-                label="🗣️User Message",
-                placeholder="Enter your question...",
-                value="How should I explain the Internet?"
-            )
-            
-            with gr.Row():
-                max_tokens = gr.Slider(
-                    minimum=1,
-                    maximum=512,
-                    value=128,
-                    step=1,
-                    label="🪙Maximum Tokens"
-                )
-                temperature = gr.Slider(
-                    minimum=0.1,
-                    maximum=1.0,
-                    value=0.7,
-                    step=0.1,
-                    label="🌡️Temperature"
-                )
-            
-            submit_btn = gr.Button("🚀 Generate Response")
-        
-        with gr.Column():
-            output = gr.Textbox(
-                label="🤳🏻Phi-4",
-                lines=10
-            )
-    
-    gr.Examples(
-        examples=examples,
-        inputs=[system_prompt, user_prompt, max_tokens, temperature],
-        outputs=output,
-        fn=generate_response,
-        cache_examples=False,
-        label="Example Prompts"
-    )
-    
-    submit_btn.click(
-        fn=generate_response,
-        inputs=[system_prompt, user_prompt, max_tokens, temperature],
-        outputs=output
-    )
-    
-    gr.Markdown("""
-    ### 📝 Parameters:
-    - **System Prompt**: Sets the behavior/role of the AI (e.g., medieval knight, wizard, merchant)
-    - **User Prompt**: Your question or input about modern concepts
-    - **Maximum Tokens**: Controls the maximum length of the generated response
-    - **Temperature**: Controls randomness (higher = more creative, lower = more focused)
-    
-    ### 💡 Tips:
-    1. Try different historical personas in the system prompt
-    2. Ask about modern technology from a historical perspective
-    3. Adjust temperature for more varied or consistent responses
-    4. Use the examples below for inspiration
-    """)
-
-# Launch the demo
-if __name__ == "__main__":
-    demo.launch(ssr_mode=False)
+gr.load(
+name='transformers:phi-4',    src=ai_gradio.registry).launch()