Update app.py

app.py

@@ -6,13 +6,22 @@ client = InferenceClient("google/gemma-3-27b-it") #change the LLM "HuggingFaceH4
 
 def respond(message, history):
 
-    messages = [{"role": "system", "content": "You are a mean and sassy office worker"}] # "content" is where u can change the personality
+    messages = [{"role": "system", "content": "You are the goofy nerd in highschool"}] # "content" is where u can change the personality
     if history:
         messages.extend(history)
         
     messages.append({"role" : "user", "content" : message})
-
-    response = client.chat_completion(messages, max_tokens = 130) #max_tokens is length
+    
+    response = ""
+    for messages in client.chat_completion(
+        messages, 
+        max_tokens = 130,
+        stream=True,
+                          ):
+        token = messages. choices[0].delta.content
+        response += token
+        yield response 
+        #max_tokens is length
     
     print(response["choices"][0]["message"]["content"].strip())