src/streamlit_app.py

import streamlit as st
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

st.set_page_config(page_title="Qwen Chatbot", layout="centered")
st.title("🧠 Qwen3-0.6B Chatbot")

@st.cache_resource
def load_model():
    tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-0.6B")
    model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-0.6B", torch_dtype=torch.float32)
    return tokenizer, model

tokenizer, model = load_model()

# Chat state
if "chat_history" not in st.session_state:
    st.session_state.chat_history = []

user_input = st.text_input("You:", key="input")

if user_input:
    # Add to chat history
    st.session_state.chat_history.append(("You", user_input))

    # Prepare prompt with full context
    context = ""
    for speaker, msg in st.session_state.chat_history:
        context += f"{speaker}: {msg}\n"
    context += "Bot:"

    inputs = tokenizer(context, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=100, do_sample=True, top_p=0.9, temperature=0.7)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)

    # Extract only bot response (after "Bot:")
    bot_msg = response.split("Bot:")[-1].strip()
    st.session_state.chat_history.append(("Bot", bot_msg))

# Display conversation
for speaker, msg in st.session_state.chat_history:
    st.markdown(f"**{speaker}:** {msg}")