Luma AI's Ray3 Video Model Now Available in Adobe Firefly

Sep 19, 2025
Luma AI's Ray3 video model is now integrated into Adobe Firefly, offering users advanced video generation capabilities. The model supports HDR and provides enhanced realism and control.

Luma AI's latest video model, Ray3, is now available in the Adobe Firefly app, announced on their website. As the first launch partner for Ray3, Adobe Firefly users gain early access to this advanced model, which will be exclusive to the app and Luma AI's Dream Machine platform for the next two weeks.

Ray3 enables the generation of cinematic, high-quality videos up to 10 seconds long, with support for native High Dynamic Range (HDR). This feature delivers richer contrast and brighter highlights, making it suitable for professional-grade video production. The model is built on a new multimodal reasoning system, allowing it to better understand creative intent and deliver production-ready assets in complex scenarios.

For content creators, Ray3 can be used in Firefly's Text to Video feature to quickly generate complementary footage for various content types. Filmmakers and visual storytellers can utilize Ray3 in Firefly Boards to explore visual directions during pre-production. All creations with Ray3 can be synced to Adobe Creative Cloud for further editing.

Luma AI's Ray3 model is available now in the Firefly app, with unlimited generations offered to customers on a paid Firefly or Creative Cloud Pro plan until October 1.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief or Sales & Marketing AI Weekly.

Also, consider following us on social media:

Free newsletter

Sales & Marketing AI Weekly

News about AI tools and innovations for Sales and Marketing professionals.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.