bokeh.org

One-line Introduction

bokeh.org is the official website of the Bokeh project. Bokeh is a free, open-source Python library for interactive visualization, developed with support from the U.S.-based NumFOCUS foundation. It is mainly aimed at data scientists and developers who need to create high-performance, highly interactive charts and dashboards in web browsers. Users typically choose it because it can generate complex visualizations using pure Python code, without requiring JavaScript, and its output can be embedded directly into web pages or Jupyter Notebook.

Business Overview

The Bokeh project began around 2012 and was originally developed by engineers at Continuum Analytics, now Anaconda Inc., before being handed over to the NumFOCUS foundation for maintenance. It is a community-driven open-source project rather than a commercial company, so it does not sell traditional “plans” or “services.” Its core work is maintaining and iterating on the Bokeh library, as well as providing documentation, examples, tutorials, and community support. In terms of industry position, Bokeh sits alongside Plotly and Matplotlib as one of the three mainstream Python visualization libraries. It is especially strong in scenarios that require interactive web charts, thanks to its native support for server-side interactivity. Its users are mainly data scientists, machine learning engineers, financial analysts, and researchers, as well as small and medium-sized teams building internal data dashboards. Because it is fully open source, there is no sales team or customer tiering; all users have equal access to the code and documentation.

Who It’s For

Bokeh is especially well suited to the following user profiles: individual developers or data scientists who need to quickly share data analysis results with non-technical audiences in the form of interactive charts, such as via a web link; small teams that want to build lightweight internal monitoring dashboards without investing time in learning JavaScript frontend frameworks; and research or academic users who need to embed scalable, filterable charts in papers or presentations. Bokeh can also handle large enterprise-level applications, but it usually needs to be deployed together with web frameworks such as Flask/Django, or use Bokeh Server to handle real-time data streams. Less suitable use cases include static reports with no need for interactivity, where Matplotlib is lighter; scenarios with very demanding mobile touch interaction requirements, since Bokeh’s mobile compatibility is only average; and projects requiring advanced 3D or geospatial visualization, where Plotly or Deck.gl would be a better fit.

Key Features and Highlights

• Interactive charts in pure Python: Create zooming, panning, hover tooltips, linked clicking, and other interactive features without writing a single line of JavaScript.
• Real-time updates with Bokeh Server: Supports automatically pushing frontend updates when backend data changes, making it suitable for real-time monitoring dashboards.
• Multiple output formats: Charts can be exported as standalone HTML files, embedded in Jupyter Notebook or Flask/Django applications, and can also be generated as static SVG/PNG images, with Selenium required.
• Rich chart types: In addition to basic line charts, bar charts, and scatter plots, it includes built-in support for maps, based on Google Maps or tile maps, network graphs, and high-density scatter plots accelerated with WebGL.
• Layered API design: High-level APIs such as figure() make it easy to get started quickly, while low-level APIs such as Plot and GlyphRenderer offer extensive customization.
• Active community and documentation: The official documentation includes many code examples and tutorials, the GitHub repository is actively maintained, and the community forum is responsive.

Pricing Analysis

Bokeh is a 100% free and open-source tool under the BSD license, with no fees for either personal or commercial use. Its pricing tier is therefore “zero cost,” giving it excellent value for money among Python visualization libraries. There are no hidden fees, subscription charges, or cloud service add-ons. The only indirect cost is that if users need to deploy Bokeh charts as a public web service, they must pay for their own server hosting, such as AWS or Alibaba Cloud ECS, but the Bokeh library itself is free. Compared with Plotly’s paid enterprise edition, which starts at several thousand dollars per year, and Tableau’s expensive licensing, Bokeh has a clear advantage for budget-sensitive projects.

How Chinese Users Can Use It

Network accessibility: The bokeh.org website and documentation are directly accessible from mainland China and are not blocked. When installing the Bokeh library via pip or conda, mirror sources such as Tsinghua and Alibaba Cloud are well supported and download speeds are stable. Payment methods: Since it is completely free, there is no payment process, so PayPal or Chinese credit card compatibility is not an issue. Is a VPN required? Daily development, documentation browsing, and library installation do not require one. However, if you need to access official GitHub Issues or English-language discussions on Stack Overflow, a VPN may be needed depending on your network environment. Domestic alternatives in China: Pyecharts, a Python wrapper around Baidu ECharts, is available in China and is friendlier in terms of Chinese documentation and domestic server deployment, but its interaction flexibility and backend real-time update capabilities are weaker than Bokeh’s. Alibaba Cloud’s DataV also provides enterprise-grade visualization services, but it is a paid product and is more focused on big-screen display scenarios.

Pros and Cons

Pros:

• ✅ Completely free and open source, with no usage restrictions
• ✅ Implemented in pure Python, reducing the learning cost of web frontend development
• ✅ Powerful interactive features, suitable for data exploration and presentations
• ✅ Supports real-time data stream updates via Bokeh Server
• ✅ Detailed documentation, rich examples, and an active community

Cons:

• ❌ Relatively limited Chinese-language resources; complex issues mainly require relying on English-speaking communities
• ❌ Rendering performance for large datasets is not as strong as Plotly, especially when WebGL is not enabled
• ❌ Default styling is fairly plain and needs extra configuration to achieve a commercial-grade look
• ❌ Mobile touch interaction support is incomplete, and some gestures may fail in mobile browsers
• ❌ Deploying it as a standalone web application requires some backend knowledge, such as Tornado or Flask

Comparison with Similar Products

• Plotly: Bokeh’s direct competitor, also free and open source in its community edition, but Plotly offers the commercial Dash enterprise edition and the Chart Studio cloud service. Plotly charts look better by default and have stronger 3D and map support, but backend real-time updates depend on the Dash framework, which has a steeper learning curve. Bokeh has an edge in pure-Python interactivity and lightweight deployment.
• Matplotlib: Better suited for static, publication-quality charts. It has weaker interactivity but the most mature code ecosystem. Bokeh is better for interactive scenarios, while Matplotlib is better for papers and reports.
• Pyecharts: Friendly to Chinese users, with comprehensive Chinese documentation and rich chart styles. However, its depth of interaction and real-time data push capabilities are weaker than Bokeh’s, and because it depends on ECharts underneath, it cannot be separated from the browser environment.

Final Recommendation

Bokeh is an excellent fit for teams that need to quickly build interactive data dashboards on a zero budget, for exploratory data analysis in Jupyter Notebook where dynamic results need to be shared, and for researchers who need to embed charts in web-based papers. It is less suitable for scenarios that require highly polished commercial chart aesthetics, where Plotly or ECharts may be better; full mobile touch optimization, where native frontend libraries are preferable; or advanced 3D and geospatial visualization, where Plotly or Deck.gl is stronger. Since it is completely free, all Python data visualization developers should try it first through the official examples, such as bokeh.sampledata, without needing to make any paid purchasing decision.