Update app.py

app.py

@@ -107,7 +107,6 @@ def create_gradio_interface():
         return gr.Dropdown.update(choices=model_ids, value=model_ids[0])
 
     model_id_selector = gr.Dropdown(choices=model_language, value=model_language[0], label="Model ID")
-    model_id_selector.change(update_model_ids, inputs=model_language, outputs=model_id_selector)
 
     # Set up a checkbox for enabling AWQ compression
     enable_awq = gr.Checkbox(value=False, label="Enable AWQ for Compression")
@@ -124,10 +123,6 @@ def create_gradio_interface():
         # Return the loaded model and tokenizer
         return ov_model, tok
 
-    # Connect model selection UI to load model dynamically
-    load_button = gr.Button("Load Model")
-    load_button.click(load_model_on_select, inputs=[model_language, model_id, enable_awq], outputs=[gr.Textbox(label="Model Status")])
-
     # Create the Gradio chatbot interface
     chatbot = gr.Chatbot()
 
@@ -137,14 +132,23 @@ def create_gradio_interface():
     top_k = gr.Slider(minimum=0, maximum=50, step=1, label="Top-k", value=50)
     repetition_penalty = gr.Slider(minimum=0, maximum=2, step=0.1, label="Repetition Penalty", value=1.0)
 
-    # Run the Gradio interface
-    demo = gr.Interface(
-        fn=bot,
-        inputs=[chatbot, temperature, top_p, top_k, repetition_penalty],
-        outputs=[chatbot],
-        title="OpenVINO Chatbot",
-        live=True
-    )
+    with gr.Blocks() as demo:
+        # Create the Gradio components and add them to the Blocks context
+        model_id_selector.change(update_model_ids, inputs=model_language, outputs=model_id_selector)
+        load_button = gr.Button("Load Model")
+        load_button.click(load_model_on_select, inputs=[model_language, model_id, enable_awq], outputs=[gr.Textbox(label="Model Status")])
+
+        # Set up the chatbot UI with all the required components
+        gr.Row([model_id_selector, enable_awq])  # Arrange the dropdowns and checkbox in a row
+        gr.Row([load_button])  # Add the button below the inputs
+        gr.Row([chatbot])  # Add the chatbot output
+        
+        # Parameters for generation
+        gr.Row([temperature, top_p, top_k, repetition_penalty])  # Add sliders in a row
+
+        # Define bot function and run the interface
+        demo.queue()  # This is used to queue inputs and outputs, handling concurrent generation calls
+        demo.launch(debug=True, share=True)  # For public access
 
     return demo