Describe the shot
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.

Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.
Generation mode: Text-to-Video / Image-to-Video / Reference-to-Video. Reference image: 10. Video references are supported. Only use source material you can process, and say what each reference should guide. A reference cannot guarantee identical details.
Duration: 3s / 4s / 5s / 6s / 7s / 8s / 9s / 10s / 11s / 12s / 13s / 14s / 15s / 16s / 17s / 18s / 19s / 20s / 21s / 22s / 23s / 24s / 25s / 26s / 27s / 28s / 29s / 30s. Resolution: 720p / 1080p. Aspect ratio: 16:9 / 9:16 / 1:1 / 21:9. Check the controls for the active generation mode before submitting. Watch the entire clip for continuity, unintended objects, timing and audio where available before using it.
Generation mode
Text-to-Video / Image-to-Video / Reference-to-Video
Duration
3s / 4s / 5s / 6s / 7s / 8s / 9s / 10s / 11s / 12s / 13s / 14s / 15s / 16s / 17s / 18s / 19s / 20s / 21s / 22s / 23s / 24s / 25s / 26s / 27s / 28s / 29s / 30s
Resolution
720p / 1080p
Aspect ratio
16:9 / 9:16 / 1:1 / 21:9
Reference image
10 · JPEG / PNG · 10 MB
Reference video URL
MP4 / MOV · 200 MB
Prompt (max)
8000
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.

Prepare your shot and inspect its result.
Prepare your shot and inspect its result.
Start with a subject, setting, action and camera movement. A slow orbit around a ceramic vase or a tracking shot along a coastal road gives you a clear result to assess. Text-to-video needs no uploaded image. Use short action sequences for storyboards, product studies or social-video ideas; exact lettering and complex choreography may need another iteration.
Prepare your shot and inspect its result.
In frame mode, upload a clear first image and optionally an ending image. Prepare JPG or PNG files up to 10 MB and crop the first image to your intended composition. Reference mode combines up to 10 images with one video of 2–30 seconds. Refer to them as @image_1 or @video_1 in your description. Reference video guides a new generation; this page does not directly edit the original clip. For EvoLink, the reference video must be at least 3 seconds. Upload the clip to this site before submitting.
Prepare your shot and inspect its result.
Choose an integer duration from 3 to 30 seconds and 720p or 1080p. Text and reference workflows offer 16:9, 9:16, 1:1 and 21:9. With first/last frames, the first image determines the output shape. A three-second clip suits a single movement; leave more time for a camera move or developing action. This workspace does not offer 4K output.
Prepare your shot and inspect its result.
Sign in, select text, frames or references, and describe the motion and relevant sound. Confirm your settings and estimate, then submit once and follow the task status. The model can generate native audio; a separate sound switch, voice upload, intermediate keyframe editor and reusable character library are not available here. Add exact titles, logos and final cuts in your editing software.
Prepare your shot and inspect its result.
With KIE selected, a five-second clip without video references costs 171 credits at 720p or 255 at 1080p. A video reference adds billable input time: output duration plus the greater of input duration and output duration. A five-second output with a three-second reference costs 204 credits at 720p. Changing provider, resolution or duration changes the estimate; check it before submitting.
Prepare your shot and inspect its result.
Watch the entire result and listen to the audio. Compare faces, product shapes, reference fidelity, motion continuity and unwanted text across the full timeline. Use the result download control and open the saved file before placing it in an edit. Generation time and visual accuracy vary. Use references you have permission to process and check the applicable service and model terms before publication.
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.
View example prompt 1
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.
A controlled camera move around the main subject, with natural motion and clear cinematic framing.
View example prompt 2
Create 3–30-second videos from text, first and last frames, or visual references. Choose 720p or 1080p and review motion, sound and credits before developing your next edit.
A continuous transition from the opening reference frame to the closing reference frame, with smooth motion between them.
View example prompt 3
Put the subject, action, camera, timing, and constraints in a clear order; precise prompts make complex shots easier to control.
A precise shot brief with ordered actions, camera direction, timing, lighting, and output constraints.
Supplied promotional media; the cover is an original AI illustration. Neither is a site generation test.
Kling 4.0 Video Generator FAQ
Create video drafts from text, from a first image with an optional ending image, or from visual references. Keep each brief focused on a clear subject and movement. Results require review before publication.
Yes. Frame mode accepts a first image and an optional last image in JPG or PNG, up to 10 MB each. The first image sets the output aspect ratio. Intermediate keyframes are not exposed in this workspace.
Reference mode accepts up to 10 images and one 2–30-second video. Use @image_1, @image_2 or @video_1 to identify your materials. A video reference contributes to the cost and guides new content rather than editing the source. For EvoLink, the reference video must be at least 3 seconds. Upload the clip to this site before submitting.
Choose 3–30 seconds in whole seconds, with 720p or 1080p output. The default is five seconds at 720p. There is no 4K option on this page.
Text and reference modes support 16:9, 9:16, 1:1 and 21:9. First/last-frame generation follows the first image, so prepare its crop before uploading.
For KIE, five seconds without a video reference costs 171 credits at 720p or 255 at 1080p. Video references have additional billable time. The displayed estimate follows the selected provider and settings.
Native audio can be generated. This workspace has no separate sound switch, voice upload, intermediate keyframes, character library or direct video-editing mode. Do not assume every feature on the official Kling platform is available here.
The supplied promotional video is shown as a model demonstration, not as proof of a task run on this site. The cover is an original AI promotional illustration. Review your own generated clip for your particular use.