Build an AI Chatbot with Streamlit & OpenAI

Install Dependencies

pip install streamlit openai

Create your project folder and a requirements.txt:

mkdir streamlit-chatbot
cd streamlit-chatbot
echo "streamlit
openai" > requirements.txt
touch app.py

Store Your API Key Securely

Never hardcode API keys. Create .streamlit/secrets.toml:

# .streamlit/secrets.toml
OPENAI_API_KEY = "sk-your-key-here"

Build the Chatbot (Full Code)

Here's the complete working app.py:

import streamlit as st
from openai import OpenAI

# Setup
st.set_page_config(page_title="AI Chatbot", page_icon="🤖")
client = OpenAI(api_key=st.secrets["OPENAI_API_KEY"])

st.title("🤖 AI Chatbot")
st.caption("Powered by Streamlit + OpenAI")

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

# Display existing messages
for msg in st.session_state.messages:
    if msg["role"] != "system":
        with st.chat_message(msg["role"]):
            st.markdown(msg["content"])

# Handle new input
if prompt := st.chat_input("Ask me anything..."):
    # Add user message
    st.session_state.messages.append({"role": "user", "content": prompt})
    with st.chat_message("user"):
        st.markdown(prompt)

    # Get AI response (streamed)
    with st.chat_message("assistant"):
        stream = client.chat.completions.create(
            model="gpt-4o-mini",
            messages=st.session_state.messages,
            stream=True,
        )

        response = st.write_stream(
            chunk.choices[0].delta.content or ""
            for chunk in stream
        )

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

Run it:

streamlit run app.py

That's a complete, working AI chatbot with streaming and history. Let's break it down.

Understanding the Key Parts

Session state: Streamlit reruns the script on every interaction. To keep the conversation alive, we store it in st.session_state.messages:

if "messages" not in st.session_state:
    st.session_state.messages = [
        {"role": "system", "content": "You are a helpful assistant."}
    ]

Chat UI: st.chat_message renders a bubble with an avatar. st.chat_input shows the input box at the bottom:

with st.chat_message("user"):
    st.markdown(prompt)

with st.chat_message("assistant"):
    st.markdown(response)

Streaming: st.write_stream() displays chunks as they arrive, like ChatGPT:

response = st.write_stream(
    chunk.choices[0].delta.content or ""
    for chunk in stream
)

Add a Sidebar with Controls

Give users control over the model and reset:

with st.sidebar:
    st.header("Settings")

    model = st.selectbox(
        "Model",
        ["gpt-4o-mini", "gpt-4o", "gpt-3.5-turbo"]
    )

    temperature = st.slider(
        "Temperature", 0.0, 1.0, 0.7, 0.1
    )

    if st.button("Clear chat"):
        st.session_state.messages = [
            {"role": "system", "content": "You are a helpful assistant."}
        ]
        st.rerun()

Pass model and temperature to the API call. This instantly makes your app feel professional.

Deploy to Streamlit Cloud (Free)

1. Push your code to a GitHub repo (make sure secrets aren't committed).
2. Go to share.streamlit.io and connect your repo.
3. In the Streamlit Cloud dashboard, go to Settings → Secrets.
4. Paste your OPENAI_API_KEY in TOML format.
5. Click Deploy.

You now have a live AI chatbot with a public URL — hosted for free.