change for AI thinking

app.py

@@ -25,8 +25,10 @@ def respond(
 
     messages.append({"role": "user", "content": message})
 
-    response = ""
-    thinking_process = []
+    # Track generation state
+    full_response = ""
+    is_complete = False
+    generation_complete = False
     
     # Use chat completion instead of text generation
     for message in client.chat_completion(
@@ -38,8 +40,20 @@ def respond(
     ):
         token = message.choices[0].delta.content
         if token:
-            response += token
-        yield response
+            full_response += token
+        
+        # Check if generation is complete
+        if message.choices[0].finish_reason == "stop" or not token:
+            generation_complete = True
+            
+        # Only yield the final complete response
+        if generation_complete and not is_complete:
+            is_complete = True
+            yield full_response
+            
+    # Ensure we always yield the final response
+    if not is_complete:
+        yield full_response
 
 
 # Custom CSS for Plant Wisdom.AI styling
@@ -77,6 +91,7 @@ custom_css = """
     margin-right: 15%;
     border: 1px solid #e0ede0;
     color: #2c3338;
+    transition: opacity 0.5s ease;
 }
 
 .submit-btn {