One click, whole workflow
Recipes are here. Package a whole workflow and run it in one click
Explore recipesOne click, whole workflow
Recipes are here. Package a whole workflow and run it in one click
Explore recipesbyDecart
Animate static visual concepts into fluid cinematic video sequences in seconds for rapid style scouting and motion pre-visualization
Move from a single reference still to a finished cinematic motion draft in three frictionless steps.
Built on an accelerated diffusion transformer architecture, Lucy combines sub-ten-second rendering speeds with precise camera control and reference identity preservation.
Explore how creators animate concept art, fashion stills, product prototypes, and cinematic plates into fluid motion sequences.
From rapid storyboard pitching to lookbook motion tests, discover how creative teams integrate near-real-time video generation into their production pipelines.
Lucy generates 5-to-10 second video clips in approximately 6 to 10 seconds, achieving faster than real-time generation speeds. This near-instant turnaround enables rapid motion sketching, iterative storyboarding, and interactive visual brainstorming without waiting on batch render queues.
Lucy accepts a reference image alongside a guiding text prompt, supporting both 16:9 landscape and 9:16 vertical aspect ratios. For optimal pixel mapping, upload clean, pre-graded source images (such as 1280x720 or 720x1280) and match the aspect ratio setting to your source frame's native proportions.
Lucy excels at rapid motion prototyping, cinematic B-roll drafting, animating static character portraits, and A/B pre-visualization for commercial storyboards. Its core strength lies in preserving the composition, lighting, and subject identity of the source image while introducing fluid physical movement.
Lucy is not designed for final-cut VFX production requiring pixel-perfect hand anatomy or legible typography, as high-speed diffusion can introduce finger smearing and text distortion. It is also not suited for long-form narrative consistency across multi-scene sequences longer than 10 seconds.
Lucy was developed by Decart and released in early 2026. It is built upon an accelerated Wan-2.2 Diffusion Transformer (DiT) architecture paired with Decart's high-efficiency inference stack.
Animate static visual concepts into fluid cinematic video sequences in seconds for rapid style scouting and motion pre-visualization