use openai_embedding

app.py

@@ -11,7 +11,7 @@ from langchain.prompts import PromptTemplate
 from langchain.chains import ConversationalRetrievalChain
 from langchain_community.vectorstores import Qdrant
 from langchain.memory import ConversationBufferMemory
-from langchain_huggingface import HuggingFaceEmbeddings
+from langchain.retrievers.multi_query import MultiQueryRetriever
 
 import chainlit as cl
 
@@ -34,7 +34,7 @@ text_splitter = RecursiveCharacterTextSplitter(
 
 docs = text_splitter.split_documents(documents)
 
-embedding = HuggingFaceEmbeddings(model_name="XicoC/midterm-finetuned-arctic")
+embedding = OpenAIEmbeddings(model="text-embedding-3-small")
 
 vectorstore = Qdrant.from_documents(
     documents=docs,
@@ -79,6 +79,11 @@ llm = ChatOpenAI(
     streaming=True,
 )
 
+retriever_llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
+multiquery_retriever = MultiQueryRetriever.from_llm(
+    retriever=retriever, llm=retriever_llm
+)
+
 
 @cl.on_chat_start
 async def start_chat():
@@ -88,7 +93,7 @@ async def start_chat():
 
     qa = ConversationalRetrievalChain.from_llm(
         llm,
-        retriever=retriever,
+        retriever=multiquery_retriever,
         memory=memory,
         combine_docs_chain_kwargs={"prompt": PROMPT},
         return_source_documents=True,