Update app.py

app.py

@@ -1,5 +1,6 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+from langchain.llms import HuggingFaceInferenceClient
 from langchain_community.vectorstores.faiss import FAISS
 from langchain.chains import RetrievalQA
 
@@ -9,8 +10,11 @@ vector_store = FAISS.load_local("db.index", embeddings=None, allow_dangerous_des
 # Load the model using InferenceClient
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+# Create an LLM object from the InferenceClient
+llm = HuggingFaceInferenceClient(client)
+
 # Initialize the RetrievalQA chain
-qa = RetrievalQA.from_chain_type(client=client, chain_type="stuff", retriever=vector_store.as_retriever())
+qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=vector_store.as_retriever())
 
 def respond(
     message,