Update chat.py

chat.py

@@ -10,6 +10,8 @@ load_dotenv(dotenv_path=env_path)
 # Use Qwen 2.5 - works well with chat and is less restricted
 MODEL_ID = "NousResearch/Hermes-2-Pro-Llama-3-8B"
 HF_TOKEN = os.getenv("HF_TOKEN")
+if HF_TOKEN:
+    HF_TOKEN = HF_TOKEN.strip()
 
 def get_chat_response(message: str, history: list):
     client = InferenceClient(model=MODEL_ID, token=HF_TOKEN)