Build a Chat App with Memory in Streamlit

What You Will Build

A chat interface that:

accepts user input
shows assistant replies
remembers previous messages
persists chat history across reruns

Why Chat History is Needed

Streamlit reruns the script every time the user interacts.

Without storing messages:

chat disappears after each message
conversation context is lost

👉 Solution → st.session_state

Step 1: Install Streamlit

Bash

pip install streamlit

Step 2: Create App File

Create:

chat_app.py

Step 3: Full Working Chat App

Python

import streamlit as st
import openai

st.title("Chat App")

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

# Display previous messages
for msg in st.session_state.messages:
    with st.chat_message(msg["role"]):
        st.write(msg["content"])

# Add Clear Chat Button
with st.sidebar:
    if st.button("Clear Chat"):
        st.session_state.messages = []
        st.rerun()

# User input
prompt = st.chat_input("Type your message")

if prompt:
    # Save user message
    st.session_state.messages.append({"role": "user", "content": prompt})

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

    # Bot response
    response = openai.ChatCompletion.create(
        model="gpt-4o-mini",
        messages=st.session_state.messages
    )

    reply = response.choices[0].message.content

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

    with st.chat_message("assistant"):
        st.write(reply)

Step 4: Run the App

Bash

streamlit run chat_app.py

How This Works Internally

Streamlit Execution Model

Every interaction:

Script reruns from top
UI redraws
Session state preserves data

Why st.session_state is Critical

Without session state:

User sends message → app reruns → message disappears

With session state:

Message stored → script reruns → messages restored

Understanding Each Component

1️⃣ Initialize Session State

if "messages" not in st.session_state:
    st.session_state.messages = []

Creates persistent storage
Runs only once

2️⃣ Display Stored Messages

for msg in st.session_state.messages:

Each message is stored as:

{"role": "user", "content": "Hello"}

Roles

user → message from user
assistant → bot reply

3️⃣ st.chat_message()

Creates chat bubble UI.

Available roles:

"user"
"assistant"
"system" (optional)

4️⃣ Chat Input Widget

prompt = st.chat_input("Type your message")

Fixed at bottom
Triggers rerun on submit
Returns entered text

5️⃣ Save Messages

st.session_state.messages.append({...})

Persists conversation
Builds chat history

Previous Next