yield response

app.py

@@ -6,19 +6,24 @@ from huggingface_hub import InferenceClient
 #AI API being used
 client= InferenceClient("Qwen/Qwen2.5-7B-Instruct-1M")
 
+response=""
+
 #defining role of AI and user
 def respond(message,history):
-
-    messages = [{"role": "system", "content": "You are acting like a comforting, guiding parent helping their child navigate academia.."}]
+    system_message: "You are acting like a comforting, guiding parent helping their child navigate academia."
+    messages = [{"role": "system", "content": system_message}]
 
     if history:
         messages.extend(history) #keep adding history 
 
     messages.append({"role":"user", "content": message})
 
-    response=client.chat_completion(messages, temperature=0.8, max_tokens=300) #capping how many words the LLM is allowed to generate as a respond (300 words)
+    response=client.chat_completion(messages, temperature=0.8, stream=True, max_tokens=300) #capping how many words the LLM is allowed to generate as a respond (300 words)
 
-    return response['choices'][0]['message']['content'].strip() #storing value of response in a readable format to display
+    for message in client.chat_completion():
+        token=message.choices[0].delta.content
+        response+=token
+    yield response['choices'][0]['message']['content'].strip() #storing value of response in a readable format to display
     
 
 #Defining chatbot giving user a UI to interact, see their conversation history, and see new messages using built in gr feature