Luma adds Kling 4.0 for fast action and bigger camera moves
Kling 4.0 now sits alongside Luma’s other creative models, aimed at fast action and bigger camera moves.
Every Luma AI update in one feed — release notes, features and changes, straight from the source.
14 updates
Kling 4.0 now sits alongside Luma’s other creative models, aimed at fast action and bigger camera moves.
Give Luma an image and up to four style references, and Style Match applies the art direction while keeping your product and composition in place.
Upload one short take and Luma re-shoots the same action as performance-led, action-led or spatial-led angles you can cut together in the edit.
Luma Layers now extracts baked-in text as editable live text, matching it to the closest library font.
Luma Agents gain GPT-Image 2.5 with two modes, Flare for speed and Sunburst for precise reference-based edits that carry into video.
Luma Agents can now drive Seedance 2.5 with up to 50 text, image, video and audio references per generation.
Extracted layers work on generated or uploaded images, letting one element change while the rest stays untouched.
The lighter, faster model is meant for storyboarding and testing directions, not final-quality output.
Users can now pull files from those accounts into any Luma board on demand.
The video model joins Luma AI's existing agent lineup with no change to the point-review-continue workflow.
The Kling Omni model joins Luma AI's Agents lineup without changing the existing workflow.
The update sharpens style transfer from reference images and improves single-element edits, Luma AI says.
The model targets stronger first-pass results for fashion, architecture, manga and cinematic styles.
Innovative Dreams pairs actors and artists with real-time generative AI, though no projects or timeline were announced.