Text-led motion
Write motion causality before camera
Tram, rails, steam and reflections share one direction and timeline while the camera preserves a readable relation.
Kling V1
Start from text, one first frame or two boundary images to make a focused foundational AI shot.
Original workflow
Three starting points for the first Kling generation
Kuaishou launched Kling in June 2024 around text-led spatiotemporal motion and physical-world simulation, then announced image-to-video and boundary-frame control in July. Flux Art currently organizes those starts as text, first frame and a fixed five-second boundary-frame workflow.
Text to videoFirst-frame videoFive-second boundary framesStandard and ProAdherence and negatives
First-frame physics
Let gravity and rotation explain deformation
The frame locks artisan, glass and furnace while the prompt owns rotation, sag, contact and camera push.
Five-second boundary
Use two states to review the full path
Two images fix the flow's start and finish while the five-second brief solves channel path, eddies and stop.
Essential inputs
Decide which input owns the information
The input route defines existing visual facts. The prompt should supply motion and change instead of competing with the source.
Text to video
Build subject, space, action, camera and timing from text.
First frame
Add motion after one established opening in Standard or Pro.
Five-second start and end
Generate a fixed five-second transition between known states in Standard or Pro.
Core capabilities
Where Kling V1 fits
Use it to validate one causal motion, a basic camera move or clear start and end states rather than packing multiple shots, audio and many characters into one clip.
Build a full shot from text
Organize the brief by subject, action, environment, camera, timing and protected details so every visible change has a cause.
Add motion after a first frame
Let the image own identity, composition and material while the prompt supplies later action, force and camera relation.
Lock a five-second boundary result
Two images define start and finish while the prompt solves path, speed, continuity and stopping behavior.
Control adherence and exclusions
Adherence balances written constraints; negative prompts exclude new subjects, structural drift, flicker or unwanted camera motion.
Shot notes
Kling V1 director briefs
Each example belongs to text, one opening frame or two boundary frames so inputs do not redefine the same facts.
Tram through bridge steam
On a rainy night, one dark-green tram travels left to right into pale bridge steam. A low camera beside the right rail tracks laterally; wheels vibrate over joints, wet steel reflects the lamps, and steam parts at the nose then closes behind. Preserve one tram, rail direction and night exposure.
Molten glass under gravity
The first frame owns the glassblower, pipe, amber glass and furnace light. He rotates steadily as the molten form sags under gravity while remaining axial; a tool briefly touches the rim and withdraws. Fixed medium-close with a slight push. Preserve one person, two hands, color and furnace position.
Water through a stone channel
The first frame has water entering the upper stone channel; the last has the same flow reaching the lower square pool. Over five seconds it advances through the continuous path, forming small eddies at two turns; a high three-quarter camera tracks lower-right and stops at the final frame. Preserve bowl, channel, moss and flow direction.
Generation check
How to use Kling V1 in Flux Art
The launch-era claims of 1080p, 30 fps and up to two minutes describe the 2024 Kling product, not the current Flux Art contract. This page follows the live configuration: text and first frame support Standard or Pro; boundary frames support Standard or Pro at a fixed five seconds; all three expose adherence and negative prompts.
Create with Kling V1- Choose the information sourceUse text without an established visual, one frame for a known opening, and two frames only when the ending must be exact.
- Write one reviewable action chainOrder trigger, travel, contact, response and stop, then state camera position and tracking direction.
- Choose a mode and review by layerCheck subject structure first, then physics and path, followed by camera, boundary alignment and unwanted content.
Source notes
Official sources and page information
Model capabilities are summarized from public Kling AI and Kuaishou materials. Available modes, parameters and pricing follow the current Flux Art workspace.
Published
Compare later Kling video models
- Kling V1.5Turn an established visual direction into a controlled short clip with one or two boundary images.
- Kling V1.6Control action, timing and camera from text, a first frame, a last frame or two boundary images.
- Kling V2 MasterUse a premium route to control action, semantics, cinematic light and explicit artistic style.
- Kling V1 ImageCreate from text or direct a 1K edit with one subject or face reference and adjustable strength.
Kling V1 FAQ
What is Kling V1?
Kling V1 is the original Kling video generation model Kuaishou launched in June 2024. Its initial official description emphasized text-led spatiotemporal motion and physical-world simulation.
Which Kling V1 workflows are available in Flux Art?
Flux Art exposes text-to-video, first-frame image-to-video and start-and-end-frame video.
Which modes are available?
All three current workflows support Standard and Pro modes.
How long is the start-and-end-frame workflow?
The current Flux Art configuration fixes Kling V1 boundary-frame generation at five seconds.
Does Kling V1 support adherence and negative prompts?
Yes. Text, first-frame and boundary-frame workflows expose prompt-adherence and negative-prompt controls.
How should a text-to-video prompt be written?
Define one subject, one causal action chain, the environment, camera position and protected details in chronological order.
How should a first-frame prompt be written?
Let the image define identity, composition, material and light; describe only later action, force and camera movement.
How should start and end frames be used?
Let the two images own the boundary states and use the prompt only for path, speed, physical response, continuity and the final stop.
Can Kling V1 results be used commercially?
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.