app.py

import streamlit as st
from huggingface_hub import InferenceClient
import random


st.title("ChatGPT-like clone")

#model_name = "unsloth/Llama-3.2-1B-Instruct"
model_name = "HuggingFaceH4/zephyr-7b-beta"
client = InferenceClient(model_name)

# Session state to hold chat history
if "messages" not in st.session_state:
    st.session_state.messages = [{"role": "system", "content": "You are a helpful assistant."}]

# Display the chat history
for message in st.session_state.messages:
    if message["role"] == "user":
        st.markdown(f"**You:** {message['content']}")
    else:
        st.markdown(f"**Zephyr 7B:** {message['content']}")

# User input
user_input = st.text_input("Type your message here...", key="user_input")

# On submit
if st.button("Send"):
    if user_input.strip():
        # Add user's message to the chat history
        st.session_state.messages.append({"role": "user", "content": user_input})

        # Query the model
        with st.spinner("Zephyr is thinking..."):
            response = client.text_generation(user_input)
            #assistant_reply = response.get("generated_text", "").strip()
            assistant_reply = response.strip()

        # Add the model's reply to the chat history
        st.session_state.messages.append({"role": "assistant", "content": assistant_reply})

        # Clear the input box
        st.session_state.user_input = ""

# Instructions for the user
st.sidebar.title("Instructions")
st.sidebar.info(
    """
    - Type your message in the input box and press "Send."
    - Responses are powered by the HuggingFace `zephyr-7b-beta` model.
    """
)