Hyvä vs Luma in 2026: Speed, Cost and Extensions, Honestly Compared

Hyvä vs Luma in 2026 is no longer a bet on a young theme against the default. Hyvä Theme has been free and open source since 10 November 2025, and Luma is the frontend Adobe ships with Magento but has not redesigned in years. The question is not which one is faster, because that has been settled on every phone that opened a Magento store this year. The question is whether the cost of moving is worth it for your store, and the honest answer depends on your extensions, your checkout and your team.

What Luma is, and what Hyvä replaces

Luma is the reference theme that has shipped with Magento 2 since the first 2.0 release. It is built on RequireJS for loading scripts, Knockout.js for rendering the minicart, the customer section and the checkout, jQuery and the Magento UI components for almost everything else, and LESS for the styles. Every extension on the market is written against it first, which is Luma's single greatest strength and the reason it is still on most stores.

Hyvä is a theme built on Tailwind CSS and Alpine.js. It keeps Magento's backend, layout XML and PHP templates, and replaces the frontend stack underneath them: no RequireJS, no Knockout, no jQuery, no UI components on the storefront. A page ships the HTML the server rendered, a stylesheet purged down to the classes the templates use, and a small amount of JavaScript, instead of the hundreds of requests a Luma page assembles in the browser.

Hyvä vs Luma, category by category

Speed and Core Web Vitals

Google measures a store on three numbers from real visitors: Largest Contentful Paint inside 2.5 seconds, Interaction to Next Paint inside 200 milliseconds and Cumulative Layout Shift under 0.1, each read at the 75th percentile of page loads, with Interaction to Next Paint having replaced First Input Delay in 2024 as the responsiveness metric, as Google's Web Vitals documentation sets out.

Luma struggles with two of the three by construction. Largest Contentful Paint waits on a render-blocking stack: RequireJS resolves dozens of modules before the page is usable, and the stylesheet carries every rule the theme has ever needed. Interaction to Next Paint suffers because Knockout renders the minicart, the customer menu and the product options after the page loads, so the main thread is busy exactly when the shopper taps. Layout shift follows the same late rendering. A disciplined team can tune Luma a long way, and the server work described below helps it most, but a Luma store rarely passes all three on a mid-range phone, and that is where Google reads the numbers.

Hyvä removes the mechanism rather than tuning it. The HTML arrives rendered, the stylesheet is a fraction of Luma's, and Alpine.js adds behaviour without a framework boot. The result is not magic, and a Hyvä store on a slow server still fails Largest Contentful Paint, because no frontend can paint what the server has not sent. With the server in order, a Hyvä store usually passes all three.

Winner. Hyvä, clearly, once the server is not the bottleneck.

Cost and licensing

Here the picture changed in November 2025. Hyvä announced that it was relicensing the Hyvä Theme under OSL 3.0 and AFL 3.0, freely available on GitHub from 10 November 2025, with the modules that support the theme relicensed under OSL 3.0 alongside it, while Checkout, Enterprise and Commerce remain proprietary under its commercial licence, as Hyvä's own announcement states. Hyvä UI, the library of ready-made sections, became a separate paid product at the same time.

Luma costs nothing because it is part of Magento. So in 2026 neither theme carries a licence fee, and the comparison moves to where the money actually goes: the build and the extensions. A Luma store that exists today has already paid for its frontend. Moving it to Hyvä is a frontend rebuild: templates rewritten in Tailwind, every custom block redone, every extension checked for a Hyvä version, and the whole thing tested across the catalogue. That is weeks of work on a simple store and months on a complex one, and it is the real price of Hyvä.

The counterweight is what Luma costs to keep. Every new feature on Luma is written against a stack that fights it, every performance fix is temporary, and a store that fails Core Web Vitals pays in rankings and in conversion every month it stays red.

Winner. Luma on the day-one bill, honestly. Hyvä over the years that follow, if speed is costing you sales.

Extensions and compatibility

Every Magento extension works on Luma, because Luma is what the extension was built on.

On Hyvä, an extension that touches the storefront needs either a Hyvä version from its vendor or a compatibility module that re-implements its templates without Knockout and jQuery. Backend-only extensions, anything that works in the admin, the order flow or an integration, are unaffected. The catalogue of extensions with a Hyvä version has grown every year, and most of the well-known vendors now ship one, but a store with forty extensions will still find a handful with none. Each of those is a decision: port it, replace it, or drop it.

This is the part of a migration that decides the estimate, which is why it is audited before anything is quoted. The audit lists every storefront extension, says which has a Hyvä version, which has a compatibility module and which has neither, and the price follows from that list.

Winner. Luma. It works with everything, and that will remain true for as long as Luma is the default.

Development and maintenance

A developer working on Luma needs RequireJS, Knockout, jQuery, the UI component framework, LESS and the layout system, and most of the hard bugs live where two of those meet. A developer working on Hyvä needs Tailwind, Alpine.js and the same layout system, the templates are plain PHP that render plain HTML, and a change that took a day on Luma often takes an hour.

The caveat is hiring. Luma developers are everywhere, because every Magento developer since 2015 has worked on it. Hyvä developers are fewer, though the gap has narrowed quickly since the theme went open source. If your in-house team has never touched Tailwind, budget time for them to learn it, and ask any agency you consider how many Hyvä stores it has shipped rather than whether it can.

Winner. Hyvä for the work itself, with a note to check who will do it.

Checkout

This one surprises people. Moving to Hyvä does not force a checkout change: the Hyvä Theme runs with Magento's default checkout, which is the same Knockout-based checkout Luma uses, so the one page a merchant worries most about can stay exactly as it is. Hyvä Checkout, the faster replacement written in the Hyvä stack, is a separate product under the commercial licence, and whether to add it is a decision made on its own numbers after the theme is live.

In practice the default checkout is the slowest page on a Hyvä store, precisely because it still carries the Luma stack. Stores that make the move usually do it in two steps: the theme first, the checkout once the first step has paid for itself.

Winner. A draw. Both run the same default checkout, and only Hyvä has an upgrade path.

Upgrades and support

Luma upgrades with Magento, because it is Magento. Hyvä tracks Magento releases on its own schedule: its documentation lists Magento 2.4.4-p9, 2.4.5-p8, 2.4.6-p7, 2.4.7-p1 and higher as supported, and PHP 8.1 through 8.4, with older PHP no longer supported. The current Hyvä releases depend on the content security policy nonce provider that arrived in Magento 2.4.4, so a store on an older line needs the Magento upgrade before the theme.

That order matters more than it sounds. A store on 2.4.3 or earlier has two projects ahead of it, and the upgrade comes first. A store already on a supported line can do either in either order, and doing the theme during the upgrade saves a round of regression testing.

Winner. Luma if you are stuck on an old version. A draw once you are current.

The comparison at a glance

Which one to choose

Choose Luma if

• Your store already passes Core Web Vitals, or your traffic is overwhelmingly desktop and the phone numbers do not move revenue.
• You run a long tail of extensions with no Hyvä versions and no budget to replace them.
• You are on Magento 2.4.3 or earlier and the upgrade has to come first; moving the theme now would mean doing the work twice.
• Your frontend is a year or two from a redesign anyway, and you would rather do the move and the redesign together.

Choose Hyvä if

• Search Console shows the store red on Core Web Vitals and the server has already been tuned.
• Your catalogue and checkout are standard enough that the extension audit comes back clean, or the extensions that matter have Hyvä versions.
• Your team ships frontend changes every week and the cost of doing so on Luma is what hurts.
• You are planning a redesign and would rather not build it on a stack you intend to leave.

How to decide with numbers

Start with the baseline. The free Magento audit reads speed, Core Web Vitals, SEO and security from the storefront without any access, and it says whether the theme or the server is the thing holding the store back.

Then take the server seriously before the theme. The approach on our Magento speed optimization page is server first: full-page cache and Varnish made to hit, OpenSearch and the database tuned, and only then the frontend.

If the numbers say the frontend, the move itself is described on our Hyvä theme development page: the extension audit, the templates rebuilt in Tailwind, the checkout decided on its own numbers, and the rankings kept with the URLs and structured data carried across.

Key takeaways

• Hyvä wins on speed and on day-to-day development; Luma wins on extension compatibility and on the cost of staying where you are.
• Since 10 November 2025 neither theme carries a licence fee for the theme itself; the price of Hyvä is the rebuild, and the price of Luma is every month in the red.
• The default checkout stays the same on both, so the move does not have to touch the page merchants fear most.
• Decide with the audit and the server work first. The theme is the last variable, not the first.

Conclusion

Hyvä vs Luma in 2026 is a question with a date on it. Before November 2025 the honest advice to a small store was often to stay on Luma and tune it; since then the licence objection is gone, the extension catalogue has caught up with most stores, and the remaining argument for Luma is the cost of leaving it. That cost is real, and for a store that passes Core Web Vitals today it may be the right answer for years. For a store that does not, and whose server has already been fixed, the numbers now point one way.