TLDR: Luma AI has launched Ray3, a groundbreaking generative video model that introduces ‘reasoning’ capabilities and native High Dynamic Range (HDR) support. This allows for more coherent, professional-grade video generation, with initial exclusive availability through Adobe Firefly and Luma AI’s Dream Machine platform.
Palo Alto, CA – September 18, 2025 – Luma AI, a leading generative AI lab, today announced the release of Ray3, its latest and most advanced generative video model. Ray3 distinguishes itself by being the first video AI model to incorporate ‘reasoning’ capabilities, enabling it to better understand creative intent, plan coherent scenes, maintain character consistency, and produce natural-looking motion. This marks a significant leap from previous generative models, which Luma AI CEO and Co-founder Amit Jain described as ‘more like slot machines – powerful but not intelligent.’
A cornerstone feature of Ray3 is its native support for High Dynamic Range (HDR) video, making it the first model capable of producing video in true 10-, 12-, and 16-bit HDR ACES2065-1 EXR format. This professional-grade output ensures richer contrast, deeper shadows, and brighter highlights, meeting the stringent technical standards required for high-end film, advertising, and game development projects. Creators can also convert existing Standard Dynamic Range (SDR) videos into generative HDR for enhanced color and editing flexibility.
The model generates cinematic, high-quality video clips up to 10 seconds long. It also features a ‘draft mode’ for rapid iteration and experimentation, allowing users to quickly explore concepts before rendering preferred results in 4K HDR. Furthermore, Ray3 is designed to critique its own outputs and suggest revisions, and production work can be saved and resumed, offering a more integrated and efficient workflow.
Luma AI has partnered with Adobe, making the Adobe Firefly app the exclusive initial platform for Ray3, alongside Luma AI’s own Dream Machine platform, for the next two weeks. This collaboration provides Firefly customers with early access to the advanced model. Amit Jain emphasized the synergy, stating, “Ray3 was built with filmmakers, storytellers, and creators at the center, and this integration with Adobe Firefly makes that vision real. By combining Adobe’s creative app ecosystem with the intelligence of Ray3, we are giving creators the ability to move from idea to cinematic video in seconds with professional-grade quality and control. This partnership isn’t just about faster workflows; it is about unlocking entirely new ways to imagine and tell stories.”
Also Read:
- YouTube Unleashes Advanced AI Creation Tools for Shorts Creators
- PictoryAI Unveils Advanced AI Storyboard for Accelerated Visual Content Generation
Ray3 is poised to revolutionize workflows for a diverse range of professionals. Content creators can leverage it within Firefly’s Text to Video feature to quickly generate b-roll or background footage for product tutorials, Instagram Reels, or TikToks. Filmmakers, producers, and visual storytellers can utilize Ray3 in Firefly Boards for storyboarding, exploring visual directions, generating environments, and prototyping shot compositions and camera perspectives. Luma AI, backed by investors like Nvidia and Andreessen Horowitz, continues its mission to build multimodal general intelligence for generating, understanding, and operating in the physical world, with Ray3 representing a significant stride in intelligent creative work.
