app.py

from langchain.chat_models import ChatOpenAI
from langchain.prompts import ChatPromptTemplate
from langchain.schema import StrOutputParser
from langchain.schema.runnable import Runnable
from langchain.schema.runnable.config import RunnableConfig
from langchain_community.vectorstores import FAISS
from langchain_openai import OpenAIEmbeddings
from langchain_core.runnables.passthrough import RunnablePassthrough

import logging

import chainlit as cl
import os

openai_api_key = os.environ["OPENAI_API_KEY"]


@cl.on_chat_start
async def on_chat_start():

    prompt_template = """
      You're a helpful AI assistent tasked to answer the user's questions.
      You can only make conversations based on the provided context. If a response cannot be formed strictly using the context, politely say you don’t have knowledge about that topic.

      CONTEXT:
      {context}

      QUESTION: {question}

      YOUR ANSWER:"""


    embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key, model="text-embedding-3-large")
    vector_store = FAISS.load_local("./faiss_index", embeddings, allow_dangerous_deserialization=True)
    model = ChatOpenAI(api_key=openai_api_key, streaming=True)
    prompt = ChatPromptTemplate.from_messages([("system", prompt_template)])

    retriever = vector_store.as_retriever()

    runnable = {"context": retriever, "question": RunnablePassthrough()} | prompt | model | StrOutputParser()

    cl.user_session.set("runnable", runnable)


@cl.on_message
async def on_message(message: cl.Message):
    logging.info(f"""
    =================================================================================
     ON MESSAGE: {message.content}
    =================================================================================
    """)
    runnable = cl.user_session.get("runnable")

    msg = cl.Message(content="")

    res = runnable.invoke(message.content)

    await cl.Message(content=res).send()
    logging.info(f"Sending message wirh res <{res}>")
    logging.info(f"Done with <{message.content}>")