Luma turns one static ad into versions for different placements

Choose the sizes and languages, generate a batch, then check each version against the approved design.

Jordan Reid
October 5, 2026
Tools, Generative Media
2 mins

A Heidelberg printing press feeding sheets of white paper — Illustrative archival print production photograph by André Robillard, dated 2014, via Wikimedia Commons under CC0 1.0. It does not depict Luma Ad Variants, a generated output or a digital product screenshot. No endorsement is implied.

Listen to this article

An approved ad often needs several sizes and languages before a campaign can run. Luma’s Ad Variants, introduced on October 1, puts those adaptations in one workflow for static ads.

Choose the sizes and languages together

Users upload a master advertisement and select square, portrait, story, medium rectangle or widescreen formats, along with the languages they need. The tool produces a version for each selected combination.

Luma says the workflow preserves key brand elements while adapting layout and copy. Users can preview and regenerate results, then download them or send them to a Luma Canvas board. The Ad Variants card is available in Discover. This initial release covers static ads.

Luma already offered broader batch generation

Luma already offered broader variation and localization through its creative agents. Its March 9 guide to batch generation describes adapting assets across formats, markets and languages. The October announcement packages a specific advertising task into a dedicated interface. That narrower change is useful for teams that want a repeatable starting point.

Use the original design as the reference

An earlier Luma guide on reference assets recommends keeping original logos, characters and other fixed elements as the source for subsequent revisions. It warns that working only from the latest generated version can introduce unwanted changes. That guidance gives teams a practical baseline for checking a batch against the approved design.

Check the copy before the campaign runs

Before distributing the ads, check that the text is readable at each size and that product details and brand marks still match the approved design. Have someone who understands the target language review the translations. This report includes no product test. The announcement offers no independent evidence that its ads perform better.

Illustrative archival print production photograph by André Robillard, dated 2014, via Wikimedia Commons under CC0 1.0. It does not depict Luma Ad Variants, a generated output or a digital product screenshot. No endorsement is implied.

Jordan Reid

Jordan Reid is focused on AI tools, agents, developer products, and the way technology changes everyday work. Jordan approaches a launch from the user’s side of the screen. What can it actually help someone finish? The voice is practical, conversational, and skeptical of products that turn a simple job into five new settings. Coverage follows coding assistants, creative software, browser agents, and the workflows around them, with attention to pricing, permissions, setup, and the human work that remains.