app.py

from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
from langchain_nvidia_ai_endpoints import ChatNVIDIA
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
import gradio as gr
import os


prompt = ChatPromptTemplate.from_messages([("system", "You are a helpful AI assistant named Arun."), ("user", "{input}")])

llm = ChatNVIDIA(model="mistralai/mixtral-8x7b-instruct-v0.1")
chain = prompt | llm | StrOutputParser()


def chat(prompt, history):

    output = ""
    for chunk in chain.stream({"input": prompt}):
        output += chunk
        yield output

demo = gr.ChatInterface(chat, title="ArunGPT",theme = gr.themes.Soft(), description="Hello this is chatbot is created for only educational purpose and is powered by mistral 8x 7b model").queue()

demo.launch()