Update app.py

app.py

@@ -7,24 +7,33 @@ client = InferenceClient("Qwen/Qwen2.5-72B-Instruct")
 
 def respond(message, history):
     
-    messages = [{"role": "system", "content": "You are a wise 80 year old chatbot."}] #change the personality
+    messages = [{"role": "system", "content": "You are a sassy chatbot."}] #change the personality
     
     if history:
         messages.extend(history)
 
     messages.append({"role":"user", "content": message})
 
-    response = client.chat_completion(
+    response = ""
+    client.chat_completion(stream=True)
+    for message in response:
+        
+
+    response = ""
+    for message in client.chat_completion(
         messages,
         max_tokens=500,
-        temperature = 0.3,
-        top_p = 0.3
-    )
+        stream=True,
+        #temperature = 0.3,
+        #top_p = 0.3
+    ):
+        token = messages.choices[0].delta.content
+        response += token
     # change length using max_tokens
 
     print(response['choices'][0]['message']['content'].strip())
     
-    return response['choices'][0]['message']['content'].strip()
+    yield response['choices'][0]['message']['content'].strip()
         
 def random_message(message, history):
     choices = ["yes.", "no.", "it is certain","without a doubt","outlook good","ask again later", "better not tell you now","very doubtful","don't count on it","my sources say no","outlook not so good","very doubtful","reply hazy, try again", "cannot predict now"]