Update app.py

app.py

@@ -37,15 +37,9 @@ Here are relevant extract from docs that you can use to generate the answer:
 
         messages.append({"role": "user", "content": processed_prompt})
 
-        generate_kwargs = dict(
-            temperature=0.6,
-            max_tokens=8192,
-            top_p=0.95,
-        )
-
         output = ""
 
-        for token in inference_client.chat_completion(messages, stream=True, **generate_kwargs):
+        for token in inference_client.chat_completion(messages, stream=True):
             new_content = token.choices[0].delta.content
             output += new_content
             yield output + f"\n\nsources: {sourced_md}"