SStreamlit × Asyntai50 free messages

AI Chatbot for StreamlitOne line of Python. Your users get help.

How to use the app, what a chart means, where the data comes from: your AI assistant answers users in seconds from your own docs. Install the free package, call chatbot() with your widget ID, and the chat is in your app.

Start for free See the setup

No credit cardStreamlit 1.28 and newerFree pip package

atelierlila.com/products/garance-linen-dress

Atelier Lila

Garance linen dress

$98.00

XSSML

Add to cart

Atelier LilaOnline

Hi! Any questions about this dress?

5 minto install and add one line

24/7answers, weekends included

1line of Python

50free messages to get started

What it handles

The questions your customers ask every day

They come right before the purchase, often in the evening. Without a quick answer, the cart stays empty. The chatbot answers them with your store's own information.

When will it arrive?How do I return an item?Which size should I pick?Can I pay with Afterpay?Do you offer 2-day shipping?My promo code doesn't work

Where is my order?Do you ship to Canada?What fabric is it?Do you offer gift wrapping?When is it back in stock?Do you sell wholesale?

Features

Built for Streamlit apps

Works with Streamlit 1.28 and newer on Python 3.10 or newer, on your own server and on Streamlit Community Cloud.

It knows your whole catalog

Product pages, categories, terms of sale, shipping and returns pages. Add your PDFs (size guide, data sheets) for everything else.

Garance linen dress$98.00

Ondine shirt$68.00

Size chart.pdfRead by the AI

Live in 5 minutes

One pip install and one chatbot() call. No HTML component to write and no JavaScript to paste.

5min

Speaks your customer's language

A shopper from Mexico, Quebec or Germany gets an answer in their own language, automatically.

ENESFRDEPTIT

It captures contacts

Quotes, trade orders, complex questions: it asks for the email and saves it.

Claire B.Quote for 40 pieces

New

Desktop and mobile

The chat adapts to every screen and sits clear of your header, menus and footer.

See what's blocking the sale

Conversations show the topics that keep coming back. You know what to add to your product pages.

Shipping38%

Sizes26%

Returns17%

Only your content

The AI answers from your pages and documents. It doesn't make up prices, delivery times or promises, and it doesn't answer off-topic questions.

Can you write me a poem?I only answer questions about the Atelier Lila store. Can I help you choose a size?

In action

An assistant that knows your store

It answers from your product pages, suggests the right items and wears your brand's colors.

Answers drawn from your pages
Sizes, delivery times, returns: it uses your own information and makes nothing up.
It can suggest your products
It shows items from your catalog as cards, with the price and a link to the product.
It passes contacts to you
For a quote or a trade order, it collects the customer's email.

Try a color

Setup

Setup with the Python package

Asyntai has a free Streamlit package on PyPI. You install it and call chatbot() near the top of your app, and the chat bubble shows in the corner. Four steps, five minutes.

Copy your widget ID

Sign in to Asyntai and open the Install page. Your widget ID is the data-asyntai-id value in the snippet. You can also paste the whole snippet.

Install the package

Run pip install streamlit-ai-chatbot. On Streamlit Community Cloud, add streamlit-ai-chatbot to your requirements.txt.

Add one line

Import chatbot from streamlit_ai_chatbot and call chatbot("YOUR_ID") near the top of your app file.

Run, and you're live

Run streamlit run app.py, or deploy. The chat bubble is in the corner of the page and answers your users.

Call chatbot() with no ID, and the package reads ASYNTAI_WIDGET_ID from st.secrets or from an environment variable. Your code then holds no ID.

my-app app.py

app.pychatbot

1 import streamlit as st 2 from streamlit_ai_chatbot import chatbot3 4 chatbot("YOUR_ID")5 st.title("Sales dashboard")

streamlit run app.py

Why Asyntai

What Streamlit sites gain

Start free with 50 messages, then grow with your store.

See pricing Start for free

Fewer abandoned carts

A customer with a doubt right before paying gets an answer straight away. A question about shipping costs no longer makes them leave the store.

Fewer repetitive requests

Shipping, tracking, returns and sizes get an automatic answer from your own pages. You only handle the requests that really need you.

More sales, same traffic

You already pay to bring in visitors with Google Ads, Meta or marketplaces. The chatbot helps them find the right product and order.

FAQ

Frequently asked questions

Everything you need to know to add an AI chatbot to your Streamlit store.

Do I need to write an HTML component?

No. The package adds the chat with one chatbot() call. You do not write components.html code or any JavaScript.

Which versions does it support?

The package works with Streamlit 1.28 and newer, on Python 3.10 or newer.

Will it slow down my app?

No. The chat script loads asynchronously in the browser. Your Python code, your queries and your charts run as before.

How does the chatbot learn about my app?

Add your docs site, help pages or FAQ in the Asyntai dashboard. You can also upload documents, for example a data dictionary, and add question and answer pairs.

Does it work on Streamlit Community Cloud?

Yes. Add streamlit-ai-chatbot to requirements.txt, and put ASYNTAI_WIDGET_ID in the Secrets of your app if you do not want the ID in your code.

Can I use it in several apps?

Yes. Each app can have its own Asyntai site, knowledge base and tone. Use a different widget ID in each app.

Does it answer readers who speak another language?

Yes. The chatbot detects the reader's language and replies in it, even if your docs are only in English.

Can I use it on several sites?

Yes. Each site has its own setup and knowledge base. The Free plan covers 1 site, Starter 2, Standard 3 and Pro up to 20 sites.

Guide

How an AI chatbot helps Streamlit apps

Great dashboards, many questions

Streamlit is a Python framework for data apps, dashboards and AI tools. Data teams build apps that many people in the company use. Those users still ask: what does this metric mean, where does the data come from, how do I filter this.

An AI chatbot answers them inside the app. It reads your docs and the documents you add, then replies in plain words, in the tone you choose. Users get an answer at once, and your data team gets fewer messages.

Answering the questions that block an order

Most questions arrive just before the purchase: shipping times, returns if the size is wrong, 2-day shipping, paying in installments. If the shopper can't find the answer in a few seconds, they close the tab.

The chatbot answers from your shipping, returns and terms pages. It reminds them of your return window, the carriers you use and your free-shipping threshold. The shopper feels reassured and completes the order.

A salesperson who knows every product

In a store, a salesperson helps with the choice. Online, the shopper is alone with your product pages. The chatbot takes that role: it compares two models, explains the difference between materials, recommends a size or suggests a matching item from your catalog.

Because it only answers from your content, it never recommends products you don't sell and never makes up specs. The richer your product pages, the more precise the answers.

Useful insights for your next campaigns

Every conversation is saved and searchable. Over the weeks, you see which products interest people most, the objections that keep coming up and the information that is missing.

If many shoppers ask whether a product comes in another color, you know what to add to your catalog. If shipping costs keep coming up, you can show them more clearly. The chatbot becomes both a customer service tool and a source of insight to grow your store.

Ready to add an AI assistant to your Streamlit app?

Start free with 50 messages and see how many questions your customers ask every day.

Start for free See pricing

Other platforms

Flask FastAPI Django Quarto Lektor MkDocs Sphinx React Bridgetown