Finish each step before the next
Dusting, lifting, folding and placing form a clear sequence instead of happening at once.
Update your browser, then reload this page. Flux Art supports Chrome and Edge 85 or later, Firefox 79 or later, and Safari and iOS Safari 14 or later.
Reload Flux ArtCheck your network connection and reload this page. If the problem continues, clear your browser cache and try again.
Reload Flux ArtAI E-commerce product suites are now live
Upload a product image, choose a platform and the images you need, then create listing-ready main images, white-background images, selling-point images, and lifestyle scenes. Product suites currently support Taobao/Tmall, JD.com, Pinduoduo, and Douyin E-commerce, with every result saved together for easy review, refinement, and export.
Control action, timing and camera from text, a first frame, a last frame or two boundary images.
Kling V1.6 officially improved motion, temporal action, camera response and visual quality. Flux Art also exposes a last-frame route: first frame describes what follows, last frame explains how the shot arrives, and boundary frames own only the transition.
Dusting, lifting, folding and placing form a clear sequence instead of happening at once.
Water-level rear-right tracking and the post-bridge rise stay tied to the kayak.
The last frame locks lantern, family and lake composition; the prompt owns release, rise and final arrival.
The input route defines existing visual facts. The prompt should supply motion and change instead of competing with the source.
Control temporal action and camera from text.
Describe what happens after a fixed opening.
Explain how the shot arrives at a fixed ending.
Constrain the transition with two boundary states.
Use it for clear temporal action, simple camera moves or foundational shots that must depart from or arrive at a specific frame.
Order preparation, execution, contact and stop to reduce simultaneous or missing steps.
State position, tracking direction, push speed and stop point so subject-camera relation is reviewable.
Choose one or two images based on where the constraint sits, without redescribing states the image already defines.
Official improvements cover color, light, detail and natural expression; prompts should still constrain headcount and expression range.
For each frame route, write only what is missing: after, before, or between two states.
The baker dusts the table, removes dough, folds inward twice, then places it at board center and releases. Fixed medium with slight push; finish each step before the next; preserve one person, two hands and dough continuity.
The first frame owns kayak, stone bridge and green river. Alternating strokes move the boat toward the arch; camera tracks smoothly from rear-right at water level, then rises slightly after the bridge. Preserve boat, paddle, person and flow direction.
The last frame owns lantern high in stars, family on snowy shore and lake composition. Begin lower as the lantern is released, rises and drifts right; camera tilts gently and ends exactly at the frame. Preserve three people, lantern color and shore lights.
Kuaishou released Kling 1.6 in December 2024, highlighting stronger response to motion, temporal action and camera direction, smoother motion and natural expression, plus color, lighting and detail upgrades. Flux Art exposes text, first frame, Pro last frame and Pro boundary frames with Standard, Pro, adherence and negative prompts.
Create with Kling V1.6Model capabilities are summarized from public Kling AI and Kuaishou materials. Available modes, parameters and pricing follow the current Flux Art workspace.
Published
Kling V1.6 is the December 2024 model upgrade Kuaishou described as improving response to motion, temporal action and camera direction plus motion, expression, color, lighting and detail.
Flux Art exposes text-to-video, first-frame, Pro last-frame and Pro start-and-end-frame workflows.
Text and first-frame support Standard and Pro. Last-frame and start-and-end-frame routes are Pro only.
Yes. Current workflows expose negative prompts and prompt-adherence control.
Order preparation, execution, contact and stop, making each step finish before the next begins.
Specify camera position relative to the subject, tracking direction, movement speed and final stopping position.
A last frame fixes the target ending. The prompt describes how subjects and camera arrive there; the current route uses Pro mode.
Last frame constrains only the ending, while start-end frames constrain both boundary states and leave the prompt responsible for the transition.
Commercial use depends on current Flux Art and model-provider terms and your rights to prompts and reference media. Check the latest terms before client work.