Every other design tool gives you an approximation of the web. Penpot gives you the web. That's not a marketing line. It's the actual architectural decision that separates it from Figma, Sketch, and everything else in the space: Penpot's canvas is built on SVG, its layout engine uses real CSS Flexbox and Grid, and when you open the Code Inspect panel, what you see is actual HTML and CSS that a browser would render, not a translation of some proprietary format.
Kaleidos, the Spanish open-source company behind it, has been building Penpot since 2019 under an MPL 2.0 license. That means you can read the source code, contribute to it, or self-host it on your own infrastructure using Docker or Kubernetes. The files are stored as SVG and JSON, formats you own. There's no vendor lock-in because there's no vendor to lock you in.
Penpot 2.0 was a significant release. CSS Grid landed as a first-class layout option, which means you can design grid-based layouts that actually correspond to what a developer would implement. Native design tokens arrived at the same time, making Penpot the first general-purpose design tool to bake tokens into the product rather than treating them as a plugin concern. The component system was redesigned, and the UI itself got a substantial refresh. The version that exists now is materially different from the Penpot many designers wrote off in 2022.
The Code Inspect view is where the design-to-development story plays out most clearly. Select a component, open the panel, and you get the CSS that actually renders it in a browser. Not approximately. Not a suggestion. The actual rules. For developer-centric teams, or designers who write code themselves, this removes an entire category of interpretation error from the handoff process.
The self-hosting option deserves more attention than it usually gets. GDPR compliance, IP ownership, regulated industries, and teams with strict data residency requirements all have legitimate reasons to want their design files on their own servers. Penpot is essentially the only production-ready general-purpose design tool that makes this feasible without enterprise sales calls.
The honest assessment is that Penpot is not yet Figma in terms of polish or ecosystem. The plugin library is smaller. Some interface panels feel more utilitarian than considered. Real-time collaboration works well but has occasionally shown performance issues on very large files. The community is active and the pace of development has accelerated, but it's still a tool that rewards patience.
Who it's for: teams that care about IP ownership and data sovereignty. Engineers who work closely with design and want handoff specs they can trust. Organizations in regulated industries that can't use cloud-hosted design tools. Open-source believers who want their toolchain to match their values. And, increasingly, teams who have simply gotten tired of Figma's pricing trajectory.
Who it's not for: teams that live in the Figma plugin ecosystem or depend on Figma-specific features. Solo designers who want a polished, zero-setup experience. Teams where speed of getting to a shareable prototype matters more than handoff accuracy.
The cloud version is free. Paid tiers add preferred hosting region, priority support, and enterprise controls. Self-hosting is free and fully featured. The open API and MCP server mean Penpot is already starting to show up in AI-assisted design workflows.
The thing worth watching is what happens as more design work moves closer to code. Penpot is the only design tool whose underlying model is already aligned with how browsers render things. That alignment may matter more as the line between design and development continues to compress.
Latest Updates
How to integrate Penpot with your developer toolchain: APIs and webhooks for workflow automation
Turn design updates into automated workflows. Learn how to connect Penpot with GitHub Actions, GitLab CI/CD, Slack, and custom scripts using APIs, access tokens, and webhooks.
8 advanced Penpot features for product teams (and why you should use them)
Discover 8 advanced Penpot features that help product teams reduce design-to-development friction, create more efficient workflows, and maintain consistency at scale.
Set up Penpot MCP with Cursor in 5 steps and no code
The fastest and simplest way to get started with Penpot and AI workflows from scratch. Learn how to get started with Penpot and Cursor, a code editor with an integrated AI client.
Set up Penpot MCP with OpenRouter and OpenCode in 6 steps
One of the fastest ways to set up Penpot for AI workflows in just 6 steps, no coding required, and only using free, open, and/or low-cost tools.
Practical uses for math functions in design tokens
Penpot design tokens support both basic calculations and complex math functions. Let’s explore how you can use both to level up your design systems.
Every other design tool gives you an approximation of the web. Penpot gives you the web. That's not a marketing line. It's the actual architectural decision that separates it from Figma, Sketch, and everything else in the space: Penpot's canvas is built on SVG, its layout engine uses real CSS Flexbox and Grid, and when you open the Code Inspect panel, what you see is actual HTML and CSS that a browser would render, not a translation of some proprietary format.
Kaleidos, the Spanish open-source company behind it, has been building Penpot since 2019 under an MPL 2.0 license. That means you can read the source code, contribute to it, or self-host it on your own infrastructure using Docker or Kubernetes. The files are stored as SVG and JSON, formats you own. There's no vendor lock-in because there's no vendor to lock you in.
Penpot 2.0 was a significant release. CSS Grid landed as a first-class layout option, which means you can design grid-based layouts that actually correspond to what a developer would implement. Native design tokens arrived at the same time, making Penpot the first general-purpose design tool to bake tokens into the product rather than treating them as a plugin concern. The component system was redesigned, and the UI itself got a substantial refresh. The version that exists now is materially different from the Penpot many designers wrote off in 2022.
The Code Inspect view is where the design-to-development story plays out most clearly. Select a component, open the panel, and you get the CSS that actually renders it in a browser. Not approximately. Not a suggestion. The actual rules. For developer-centric teams, or designers who write code themselves, this removes an entire category of interpretation error from the handoff process.
The self-hosting option deserves more attention than it usually gets. GDPR compliance, IP ownership, regulated industries, and teams with strict data residency requirements all have legitimate reasons to want their design files on their own servers. Penpot is essentially the only production-ready general-purpose design tool that makes this feasible without enterprise sales calls.
The honest assessment is that Penpot is not yet Figma in terms of polish or ecosystem. The plugin library is smaller. Some interface panels feel more utilitarian than considered. Real-time collaboration works well but has occasionally shown performance issues on very large files. The community is active and the pace of development has accelerated, but it's still a tool that rewards patience.
Who it's for: teams that care about IP ownership and data sovereignty. Engineers who work closely with design and want handoff specs they can trust. Organizations in regulated industries that can't use cloud-hosted design tools. Open-source believers who want their toolchain to match their values. And, increasingly, teams who have simply gotten tired of Figma's pricing trajectory.
Who it's not for: teams that live in the Figma plugin ecosystem or depend on Figma-specific features. Solo designers who want a polished, zero-setup experience. Teams where speed of getting to a shareable prototype matters more than handoff accuracy.
The cloud version is free. Paid tiers add preferred hosting region, priority support, and enterprise controls. Self-hosting is free and fully featured. The open API and MCP server mean Penpot is already starting to show up in AI-assisted design workflows.
The thing worth watching is what happens as more design work moves closer to code. Penpot is the only design tool whose underlying model is already aligned with how browsers render things. That alignment may matter more as the line between design and development continues to compress.
Latest Updates
How to integrate Penpot with your developer toolchain: APIs and webhooks for workflow automation
Turn design updates into automated workflows. Learn how to connect Penpot with GitHub Actions, GitLab CI/CD, Slack, and custom scripts using APIs, access tokens, and webhooks.
8 advanced Penpot features for product teams (and why you should use them)
Discover 8 advanced Penpot features that help product teams reduce design-to-development friction, create more efficient workflows, and maintain consistency at scale.
Set up Penpot MCP with Cursor in 5 steps and no code
The fastest and simplest way to get started with Penpot and AI workflows from scratch. Learn how to get started with Penpot and Cursor, a code editor with an integrated AI client.
Set up Penpot MCP with OpenRouter and OpenCode in 6 steps
One of the fastest ways to set up Penpot for AI workflows in just 6 steps, no coding required, and only using free, open, and/or low-cost tools.
Practical uses for math functions in design tokens
Penpot design tokens support both basic calculations and complex math functions. Let’s explore how you can use both to level up your design systems.