Set max token limit to avoid errors

app.py

@@ -13,6 +13,7 @@ openai.api_key = os.getenv("OPENAI_API_KEY")
 CHAT_ENDPOINT="https://api.openai.com/v1/chat/completions"
 CHAT_MODEL = "gpt-3.5-turbo"
 CHAT_AUTH = {"Authorization": "Bearer " + openai.api_key}
+MAX_TOKENS = 4096
 gpt_history = []
 
 class ChatRoles():
@@ -24,6 +25,7 @@ def get_assistant_response():
     params = {
         "model": CHAT_MODEL,
         "messages": gpt_history,
+        "max_tokens": MAX_TOKENS
     }
     response = requests.post(url=CHAT_ENDPOINT, json=params, headers=CHAT_AUTH)
     print(literal_eval(response.content.decode("utf-8")))