Update app.py

app.py

@@ -16,6 +16,7 @@ from langchain_ollama import OllamaLLM
 from langchain.docstore.document import Document
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
+from huggingface_hub import InferenceClient
 
 os.environ["STREAMLIT_WATCHER_TYPE"] = "none"