Update app.py

app.py

@@ -2,13 +2,16 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 from langchain_community.vectorstores.faiss import FAISS
 from langchain.chains import RetrievalQA
-from langchain_community.llms import HuggingFaceHub
+from langchain_community.llms import HuggingFacePipeline
 
 # Load the vector store from the saved index files
 vector_store = FAISS.load_local("db.index", embeddings=None, allow_dangerous_deserialization=True)
 
-# Initialize the HuggingFaceHub LLM
-llm = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature": None, "top_p": None})
+# Load the model using InferenceClient
+client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+
+# Initialize the HuggingFacePipeline LLM
+llm = HuggingFacePipeline(client, model_kwargs={"temperature": None, "top_p": None})
 
 # Initialize the RetrievalQA chain
 qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=vector_store.as_retriever())