KuaishouKling 3.0 Omni Prompt Guide
How to write Kling 3.0 Omni prompts: a structure that works, 6 example prompts for text to video, image to video and reference to video, the settings that change the result and how to reference your own media. Copy any prompt into the Studio to try it.
- Example prompts
- 6
- 3 tasks
- Resolution
- 4K
- 720p · 1080p · 4K
- Clip length
- 3–15 s
- audio optional
- Per try from
- $0.0739/s
- 720p
Structure
A Kling 3.0 Omni prompt, piece by piece
Write in this order and keep each part concrete. Physical detail works better than mood words.
01
Subject
Who or what is on screen, with the details that must stay consistent: age, clothing, materials, colours.
02
Action
One clear action in the present tense. For clips up to 15 s, describe a short sequence of beats.
03
Camera
Shot size and movement: close-up, wide, dolly-in, orbit, handheld, drone.
04
Scene and light
Location, time of day, and where the light comes from.
05
Style
Lens, film look, colour grade or animation style.
06
Sound
Kling 3.0 Omni can add native audio (billed at a higher rate): describe ambience, effects or dialogue when generate_audio is on.
07
References
Say what each reference controls, e.g. @Image1: “@Image1 is the main character”.
Examples
Kling 3.0 Omni prompt examples
Prompts written for this model, grouped by task. Open one in the Studio, then change one part at a time.
- Reference to Video9:16
@Image1 tries on the jacket from @Image2 in front of a mirror, then turns to camera and smiles
Try in Studio - Text to Video16:9
A paper crane unfolds into a real bird and flies out of an open window, morning light
Try in Studio - Reference to Video16:9
@Image1 and @Image2 share an umbrella crossing a rainy Tokyo street at night, slow dolly
Try in Studio - Text to Video
A lone samurai walks through falling cherry blossoms at dusk, slow tracking shot, cinematic lighting
Try in Studio - Image to Video
She turns toward the window and smiles as the curtains sway; slow push-in
Try in Studio - Reference to Video
@Image1 walks into the café from @Image2 and sits by the window, handheld camera
Try in Studio
By task
Prompting each Kling 3.0 Omni task
Text to Video
Everything comes from the text, so describe subject, action and camera explicitly. Put the most important element first.
Image to Video
The image is the first frame: describe the motion and camera from that frame instead of re-describing the picture. Add a last frame to control where the shot ends.
Reference to Video
Attach references and mention them in the prompt by name, saying what each one should control — character, product, style or camera move.
Referencing your own media
| Input | In the prompt | Limits and rules |
|---|---|---|
| Reference images Reference to Video · up to 7 | @Image1, @Image2… | Up to 7 images for characters, products, style or scene. Refer to them in the prompt as @Image1, @Image2… |
Settings
Settings that change the result
Set these next to the prompt — in the Studio or as API parameters.
| Setting | Options | Effect |
|---|---|---|
| Aspect ratio aspect_ratio | 16:9 · 9:16 · 1:1 default 16:9 | Output frame shape. |
| Resolution resolution | 720p · 1080p · 4K default 720p | Output resolution. Higher resolutions cost more per second. Price: 720p $0.0739/s · 1080p $0.0986/s · 4K $0.3685/s |
| Duration duration | 3–15s default 5 | Length of the generated clip in seconds (3–15s). You pay per second. |
| Generate audio generate_audio | on / off default false | Generate synchronized sound, ambience and dialogue. Clips with audio are billed at a higher per-second rate. |
| Negative prompt negative_prompt | — | Things to avoid in the output. |
FAQ