Update app.py

app.py

@@ -1,4 +1,3 @@
-
 import gradio as gr
 from huggingface_hub import InferenceClient
 
@@ -30,10 +29,10 @@ def respond(
 
     for message in client.chat_completion(
         messages,
-        max_tokens=600,
+        max_tokens=max_tokens,
         stream=True,
-        temperature=0.7,
-        top_p=0.95,
+        temperature=temperature,
+        top_p=top_p,
     ):
         token = message.choices[0].delta.content
 
@@ -49,4 +48,5 @@ demo = gr.ChatInterface(
 
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()
+