Update app.py

add Env variable for the token

app.py

@@ -45,11 +45,11 @@ def get_llm():
     # This uses Hugging Face's free Inference API to run the model
     # You don't need Ollama for this!
     llm = HuggingFaceEndpoint(
-        repo_id="ibm-granite/granite-3.1-8b-instruct", # Use the latest available on API
+        repo_id="ibm-granite/granite-3.1-8b-instruct", 
         task="text-generation",
         max_new_tokens=512,
         temperature=0.1,
-        huggingfacehub_api_token="YOUR_HF_TOKEN_HERE" # Get this from settings
+        huggingfacehub_api_token=os.getenv("HF_TOKEN") #
     )
     return llm