app.py

import streamlit as st
import os
import torch
from openai import OpenAI
import numpy as np
import sys
from dotenv import load_dotenv
import random
from huggingface_hub import InferenceClient

# Load environment variables
load_dotenv()

# Constants
MAX_TOKENS = 4000
DEFAULT_TEMPERATURE = 0.5

# initialize the client

client = OpenAI(
  base_url="https://api-inference.huggingface.co/v1",
  api_key=os.environ.get('API_KEY')  # Replace with your token
)
    
# Create supported models
model_links = {
    "Meta-Llama-3.1-70B-Instruct": "meta-llama/Llama-3.1-70B-Instruct",
    "Mistral-7B-Instruct-v0.3": "mistralai/Mistral-7B-Instruct-v0.3",
    "Gemma-2-27b-it": "google/gemma-2-27b-it",
    "Falcon-7b-Instruct": "tiiuae/falcon-7b-instruct",
}


# Random dog images for error message
random_dog_images = ["broken_llama3.jpeg"]

def reset_conversation():
    '''
    Resets Conversation
    '''
    st.session_state.conversation = []
    st.session_state.messages = []
    return None

st.sidebar.button('Reset Chat', on_click=reset_conversation) #Reset button
    
def main():
    st.header('Multi-Models')


    # Sidebar for model selection and temperature
    selected_model = st.sidebar.selectbox("Select Model", list(model_links.keys()))
    temperature = st.sidebar.slider('Select a temperature value', 0.0, 1.0, DEFAULT_TEMPERATURE)

    if "prev_option" not in st.session_state:
        st.session_state.prev_option = selected_model

    if st.session_state.prev_option != selected_model:
        st.session_state.messages = []
        # st.write(f"Changed to {selected_model}")
        st.session_state.prev_option = selected_model
        reset_conversation()

    st.markdown(f'_powered_ by ***:violet[{selected_model}]***')

    # Display model info and logo
    st.sidebar.write(f"You're now chatting with **{selected_model}**")
    st.sidebar.markdown("*Generated content may be inaccurate or false.*")

    # Initialize chat history
    if "messages" not in st.session_state:
        st.session_state.messages = []

    # Display chat messages from history on app rerun
    for message in st.session_state.messages:
        with st.chat_message(message["role"]):
            st.markdown(message["content"])

    # Chat input and response
    if prompt := st.chat_input("Type message here..."):
        process_user_input(client, prompt, selected_model, temperature)

def process_user_input(client, prompt, selected_model, temperature):
    # Display user message
    with st.chat_message("user"):
        st.markdown(prompt)
    st.session_state.messages.append({"role": "user", "content": prompt})

    # Generate and display assistant response
    with st.chat_message("assistant"):
        try:
            stream = client.chat.completions.create(
                model=model_links[selected_model],
                messages=[
                    {"role": m["role"], "content": m["content"]}
                    for m in st.session_state.messages
                ],
                temperature=temperature,
                stream=True,
                max_tokens=MAX_TOKENS,
            )
            response = st.write_stream(stream)
        except Exception as e:
            handle_error(e)
            return

    st.session_state.messages.append({"role": "assistant", "content": response})

def handle_error(error):
    response = """😵‍💫 Looks like someone unplugged something!
    \n Either the model space is being updated or something is down."""
    st.write(response)
    random_dog_pick = random.choice(random_dog_images)
    st.image(random_dog_pick)
    st.write("This was the error message:")
    st.write(str(error))

if __name__ == "__main__":
    main()