Streamlit

What is Streamlit?

Streamlit is an open-source Python library for building interactive web applications out of ordinary data scripts. A developer writes a Python file that calls functions such as st.write, st.slider or st.line_chart, launches it with the streamlit run command and gets a working app in the browser, with no HTML, CSS or JavaScript to write. Whenever a user moves a slider or clicks a button, Streamlit reruns the script from top to bottom and redraws the page, an execution model that keeps apps short and easy to follow.

Caching decorators stop expensive steps, such as loading a large file or a machine learning model, from repeating on every rerun, and session state keeps values between interactions. The library ships widgets, charts, editable data tables, layout containers, multipage navigation and chat elements, and developers can add custom components written in JavaScript when the built-in set is not enough.

Streamlit apps can run anywhere Python runs. The free Streamlit Community Cloud deploys public apps straight from GitHub repositories, while customers of the Snowflake AI Data Cloud can build and host apps inside their Snowflake accounts, where access is governed by Snowflake roles. Since large language models became widespread, Streamlit has become a common way to put chat interfaces and prototypes in front of users. The project remains under active development at Snowflake, which released version 1.64 in September 2026.

Key features of Streamlit

1. 01

   Pure Python apps

   Apps are plain Python scripts, so data scientists can turn an analysis into a shareable web app without learning front-end frameworks, HTML templates or callback-based programming.

2. 02

   Rerun execution model

   Streamlit reruns the script whenever a user interacts with a widget and updates the output automatically. Fragments let part of a page rerun on its own when refreshing everything would be too slow.

3. 03

   Widgets, charts and tables

   Built-in elements cover sliders, text inputs, file uploaders, maps, native charts and interactive or editable dataframes, with support for plotting libraries such as Altair, Plotly and Matplotlib.

4. 04

   Caching and session state

   The st.cache_data and st.cache_resource decorators store the results of slow computations and shared resources such as database connections, and session state keeps variables available between reruns for each user.

5. 05

   Chat elements for LLM apps

   Chat message containers, a chat input box and streaming text output make it straightforward to build conversational interfaces on top of large language model APIs.

6. 06

   Multipage navigation

   Apps can be split into several pages with sidebar or top navigation menus, defined in code with st.Page and st.navigation.

7. 07

   Community Cloud hosting

   Streamlit Community Cloud deploys apps for free from a GitHub repository and updates them automatically when new code is pushed, which makes it popular for demos, portfolios and open-source projects.

8. 08

   Streamlit in Snowflake

   Snowflake customers can create and run Streamlit apps inside their own accounts, keeping data in place and controlling who can open each app through Snowflake’s role-based security.

Who uses Streamlit?

• Data scientists share model results and exploratory analysis with colleagues as interactive apps instead of static notebooks or slide decks.
• Machine learning engineers build internal tools for labeling data, comparing model outputs and demonstrating prototypes to stakeholders.
• Developers create chatbot and retrieval-augmented generation demos on top of large language model APIs in a few dozen lines of Python.
• Snowflake customers build internal dashboards and data-entry apps that run directly on governed company data inside their Snowflake accounts.
• Students, researchers and hobbyists publish public data projects and portfolio apps on Streamlit Community Cloud.

History of Streamlit

Streamlit was founded in 2018 by Adrien Treuille, Thiago Teixeira and Amanda Kelly, whose backgrounds included Google X and the self-driving vehicle company Zoox. The company released the library as open source on October 1, 2019, together with a $6 million seed round led by Gradient Ventures, and it spread quickly among data scientists who wanted to share their work without building a separate web front end.

Streamlit reached version 1.0 in October 2021, by which point it had raised $62 million and was preparing a commercial cloud service. In March 2022 Snowflake announced a deal to acquire the company, reported at about $800 million, and completed the purchase at the end of that month. The framework stayed open source under the Apache 2.0 license, Community Cloud continued to host public apps for free, and Streamlit in Snowflake became generally available on AWS in December 2023.

1. 2018

   Adrien Treuille, Thiago Teixeira and Amanda Kelly found Streamlit.

2. 2019

   Streamlit is released as an open-source library alongside a $6 million seed round.

3. 2021

   Streamlit 1.0 ships in October, months after a $35 million Series B round.

4. 2022

   Snowflake acquires Streamlit in a deal reported at about $800 million.

5. 2023

   Streamlit in Snowflake becomes generally available on AWS.

6. 2026

   Streamlit in Snowflake gains a generally available container runtime.

Streamlit pricing

Streamlit alternatives

Well-known alternativesPlotly Dash, Gradio, Shiny for Python, Panel and Reflex

Streamlit: frequently asked questions

Sources

1. Streamlit streamlit.io
2. Streamlit launches open-source machine learning application development framework, TechCrunch techcrunch.com
3. Snowflake announces intent to acquire Streamlit snowflake.com
4. Streamlit reaches 1.0, TechCrunch techcrunch.com
5. Streamlit repository on GitHub github.com

Last reviewed September 17, 2026. Spotted an error? Send a correction.