Implement streaming responses for LLM API

app.py

@@ -10,6 +10,7 @@ from langchain.chains import ConversationChain
 from starlette.responses import StreamingResponse # <-- NEW IMPORT
 import asyncio 
 from langchain_community.llms import HuggingFacePipeline
+import json
 
 app = FastAPI()
 # Get the Hugging Face API token from environment variables (BEST PRACTICE)