Update app.py

app.py

@@ -560,24 +560,25 @@ def chat_function(message, history, mode, progress=gr.Progress()):
     
     try:
         # Load model (this will run on GPU when GPU is allocated)
-        progress(0.05, description="Loading model on GPU...")
+        progress(0.05)
         tok, mod, dev = load_model()
         
         # Create appropriate generator
         if mode == "Quality (Slower, Better)":
             generator = QualityGenerator(tok, mod, dev)
-            progress(0.1, description="Initializing quality mode...")
+            progress(0.1)
         else:
             generator = SpeedGenerator(tok, mod, dev)
-            progress(0.1, description="Initializing speed mode...")
+            progress(0.1)
         
         # Generate response with progress callback
-        def progress_callback(pct, desc):
-            progress(pct, description=desc)
+        def progress_callback(pct, status_msg):
+            progress(pct)
+            # We'll show status in the performance display instead
         
         response, stats = generator.generate(message, progress_callback)
         
-        progress(1.0, description="Complete!")
+        progress(1.0)
         
         # Create performance info
         perf_info = f"""**⚡ Performance Stats:**