Multimodal reasoning model that generates brand-consistent images and edits
Added Jun 15, 2026
Luma Uni-1.1 is a multimodal reasoning model that understands intention, follows reference images, and generates or edits images with style and brand consistency. It supports text-to-image, image-to-image, and multi-reference generation, and ranks highly in human preference benchmarks for overall quality, style and editing, and reference-based generation.
Why: Uni-1.1 ties a reasoning model directly to pixel generation, making it unusually good at following brand references and complex creative direction in images.
Tokyo-based Sakana AI launched Sakana Marlin, a B2B research agent that runs long-horizon reasoning jobs and can produce 100-page strategy reports in about eight hours.