How to write prompts for Kling Video 4.0
Kling Video 4.0 turns a sentence or a start picture into a clip, with sound. The first two sections put Kling’s own guide in our words, and we did not test them. The last rests on one sample clip made with this model that we watched ourselves, so it is a very small sample.
Written from the provider’s documentation and from results we checked ourselves. Last checked November 8, 2026.
Make something with Kling Video 4.0The Kling Video 4.0 model page
Concrete, and in positive words
From the provider’s documentation
- Kling’s guide says to name what can be seen, such as light, textures, materials and colours, and to avoid vague labels like premium look or vintage feel.
- It says negative instructions can increase errors, so describe what should appear. It names two exclusions worth writing because the model may add them on its own: no subtitles and no background music.
- A short prompt works: the model fills in the story, the staging and the camera. A detailed prompt is followed closely, with little added by the model.
- For a complex clip it suggests four blocks: the role of each supplied picture, the overall settings (scene, style, camera tone), a sequence of events with time ranges, and the audio.
Shots, camera, start picture and speech
From the provider’s documentation
- Multi-shot is on by default for clips from 3 to 30 seconds: the model splits the shots and sets the pace itself. To control the order and the beats, write time-coded parts such as 0 to 3 s, then 3 to 6 s.
- Name camera moves directly: push-in, pull-out, pan, tracking shot, orbit, whip pan. Say handheld or gimbal if you want that feel, and describe the path and pace of the camera next to what the subject does.
- For a start picture, Kling’s wording begins by saying that the picture is the first frame, then describes what happens next. The clip’s aspect ratio follows the first picture.
- For speech, put the lines in quotation marks, in the language that will be spoken, and describe the speaker’s age, tone, speed and mood. Sound effects, ambience and music go in a separate audio part, and music can be asked for or left out.
- Kling’s own screen labels supplied pictures @Image1, @Image2 and gives each a role. Pickshot takes reference pictures its own way, so we do not know whether those labels are understood here. Say in plain words what each picture is for.
What our sample clip showed
From results we checked ourselves
- The prompt was one sentence, “This person reads a book on a train by the window, soft landscape moving outside”, with a picture of a person supplied. The clip is three seconds long.
- It shows her reading with fields moving past the window. Part-way it cuts to a close shot of her face and the book, with a page turning, although the prompt named no cut. That fits Kling’s note that multi-shot is on by default.
- We have one clip, so we cannot say how often this happens, how closely the face matches the picture, or how the sound turns out. If you need one unbroken shot, check the result and run again if it cut.
Real results and the prompts behind them
Results made with this model. The prompts are shown exactly as written, in English.
Settings of this model
- Från
- Kuaishou
- Typ
- AI video model
- Så skapar du det
- Text till video · Bild till video · Referens till video
- Längd
- 3–30s
- Upplösning
- 720p / 1080p
- Ljud
- Med ljud
- Pris
- from 264 / sec
Frågor
Why did my Kling Video 4.0 clip cut to a different shot?
Kling’s guide says multi-shot is on by default for clips from 3 to 30 seconds. Our sample cut from a medium shot to a close-up although the prompt named no cut. Kling says to write time-coded parts to control the shots. We have not tested how to keep a single shot.
Should I write “no music” or “no subtitles”?
Kling’s guide says negative instructions can increase errors, and names these two as the exclusions worth writing because the model may add them by default. We did not test them.
How do I write a prompt for a start picture?
Kling’s guide says to begin by saying the picture is the first frame, then describe what happens next. Our one sample started from a picture of a person, and the same person appeared before and after its cut. One clip is not enough to promise a likeness.
Provider documentation we read
We read these pages and wrote the points that matter in our own words. The provider may change them.
- Kling 4.0 model user guide (Kling) (kling.ai)
More prompting guides
- How to write prompts for GPT Image 2.6 FlareAI image model
- How to write prompts for GPT Image 2.6 SunburstAI image model
- How to write prompts for Nano Banana 2.1AI image model
- How to write prompts for MOSS ImageAI image model
- How to write prompts for Grok Imagine ProAI image model
- How to write prompts for Nano Banana ProAI image model
- How to write prompts for Nano Banana 2AI image model
- How to write prompts for GPT Image 2.5 FlareAI image model
- How to write prompts for IndigoAI image model
- How to write prompts for SaffronAI image model
- How to write prompts for CeladonAI image model
- How to write prompts for LapisAI image model
- How to write edit requests for UmberAI-bildredigerare
- How to write prompts for Seedance 2.5AI video model
- How to write prompts for Seedance 2.0 MiniAI video model
- How to write prompts for Kling V3AI video model
- How to write prompts for Kling O3 OmniAI video model
- How to write prompts for Veo 3.1AI video model
- How to write prompts for Veo 3.1 FastAI video model
- How to write prompts for SilkAI video model
- How to write prompts for LacquerAI video model
- How to write prompts for SatinAI video model
Namnen på modeller, företag och produkter på denna webbplats är varumärken som tillhör respektive ägare; Pickshot är inte affilierat med, sponsrat av eller godkänt av dem.