TL;DR
- Kling 3.0 Omni has four endpoints on HiAPI (text-to-video, image-to-video, reference-to-video, and video transformation), and all of them bill per second.
- Text, image, and reference-to-video without a reference video share one rate: $0.10 per second at 720p ($0.129 with sound), $0.129 at 1080p ($0.1645 with sound), and $0.479 at 4K.
- Reference-to-video with a reference video, and transformation, use a second rate: $0.143 at 720p, $0.193 at 1080p, $0.479 at 4K. A 5-second 720p clip without sound costs $0.50, and failed tasks are free.
Price table
| Endpoint | Model id | 720p / second | 1080p / second | 4K / second |
|---|---|---|---|---|
| Text-to-video | kling-3.0-omni/text-to-video | $0.10, $0.129 with sound | $0.129, $0.1645 with sound | $0.479 |
| Image-to-video | kling-3.0-omni/image-to-video | $0.10, $0.129 with sound | $0.129, $0.1645 with sound | $0.479 |
| Reference-to-video (images only) | kling-3.0-omni/reference-to-video | $0.10, $0.129 with sound | $0.129, $0.1645 with sound | $0.479 |
| Reference-to-video (with a reference video) | kling-3.0-omni/reference-to-video | $0.143 | $0.193 | $0.479 |
| Video transformation | kling-3.0-omni/transformation | $0.143 | $0.193 | $0.479 |
Live HiAPI prices, checked October 15, 2026.
How billing works:
- 4K is the same with or without sound: $0.479 per second.
- A reference video turns sound off. With
video_urlsset,audiomust befalse, so that tier has a single price. - Transformation follows the source clip's length (3–15.5 seconds) and doesn't generate sound.
- Text, image, and reference-to-video run 3–15 seconds, and
resolutiondefaults to 720p.
Cost by duration
Text, image, and reference-to-video without a reference video:
| Duration | 720p | 720p + sound | 1080p | 1080p + sound | 4K |
|---|---|---|---|---|---|
| 3 s | $0.30 | $0.387 | $0.387 | $0.4935 | $1.437 |
| 5 s | $0.50 | $0.645 | $0.645 | $0.8225 | $2.395 |
| 10 s | $1.00 | $1.29 | $1.29 | $1.645 | $4.79 |
| 15 s | $1.50 | $1.935 | $1.935 | $2.4675 | $7.185 |
Reference-to-video with a reference video, and transformation:
| Duration | 720p | 1080p | 4K |
|---|---|---|---|
| 3 s | $0.429 | $0.579 | $1.437 |
| 5 s | $0.715 | $0.965 | $2.395 |
| 10 s | $1.43 | $1.93 | $4.79 |
| 15 s | $2.145 | $2.895 | $7.185 |
$10 covers 100 seconds at 720p without sound, about twenty 5-second clips.
What you get for the money
One paper boat through three endpoints: text-to-video for the first shot, then transformation to turn the rainy night into snow.
A paper boat drifts down a rain-soaked city gutter at dusk, neon signs reflecting and rippling on the water; the camera glides low beside it as raindrops splash around. Cinematic, shallow depth of field, smooth tracking motion.
Generate with this prompt → (3 seconds, 720p)
Text-to-video, 720p, 3 seconds, no sound, $0.30.
Transform the scene into a snowy winter night: the gutter becomes a frozen stream with drifting snowflakes, the paper boat becomes a small glowing paper lantern boat, keep the exact camera motion and water movement from the original video. Cinematic, cold blue tones.
Generate with this prompt → (upload your source video, 720p)
Transformation, 720p, follows the 3-second source, $0.429.
Omni vs the other Kling 3.0 lines
| Model | 720p / s | 720p + sound | 1080p / s | 1080p + sound | 4K / s |
|---|---|---|---|---|---|
| Kling 3.0 Omni | $0.10 | $0.129 | $0.129 | $0.1645 | $0.479 |
| Kling 3.0 | $0.10 | $0.143 | $0.129 | $0.193 | $0.479 |
| Kling 3.0 Turbo | $0.13 | no sound option | $0.16 | no sound option | — |
Without sound, Omni and standard Kling 3.0 cost the same. With sound, Omni is cheaper per second ($0.014 less at 720p, $0.0285 less at 1080p). What Omni adds is reference-to-video and transformation; for when to pick which, see Kling 3.0 Omni vs Kling 3.0.
Ways to spend less
- Iterate at 720p, 3 seconds: $0.30 a try. Move to 1080p or 4K once the prompt works.
- Leave sound off unless you need it:
audiodefaults tofalse; turning it on adds $0.029 per second at 720p. - Trim reference videos: the reference-video tier starts at $0.143 per second.
- Set
resolutionon purpose: the default is 720p; ask for 1080p explicitly.
FAQ
Does the number of reference images change the price?
No. Reference-to-video pricing depends only on resolution, sound, and whether you send a reference video.
How is transformation billed?
Per second of output, and the output length follows the source video (3–15.5 seconds).
Am I charged for failed tasks?
No. You pay only for videos that generate successfully.
Can I get 4K with sound?
Yes. 4K is $0.479 per second either way.
Integration code is in the Kling 3.0 Omni API guide, and ready-made prompts in Kling 3.0 Omni prompt examples. Or generate your first clip on the model page.
