Update app.py

app.py

@@ -11,7 +11,7 @@ FIXED_MAX_TOKENS = 99999 # Note: This is a very high value, typical values are m
 # --- Initialize the InferenceClient ---
 # For custom OpenAI-compatible APIs, initialize the InferenceClient with the base URL.
 # The specific model will be specified in the API call (e.g., chat_completion).
-API_BASE_URL = "https://vulture-awake-probably.ngrok-free.app/v1/chat/completions" # Base URL for the custom API
+API_BASE_URL = "https://epwns5mzloen5mu7.us-east-1.aws.endpoints.huggingface.cloud" # Base URL for the custom API
 
 try:
     # Initialize the client with the base URL of your API.