Start free
New

Kling O3 ReferenceAI video generator

A video that keeps the people, products or style of up to four reference images, named in the prompt as @Image1 to @Image4, with an optional first and last frame and sound. On Leap, it sits beside 169 other models in one studio.

  • Image to video
  • Up to 15 s
  • With sound
  • 3 aspect ratios
  • Start free, no card
  • Every top model, one account
  • Pay per run, priced before it starts

AI video generator

Kling O3 Reference at a glance

Kling O3 Reference is a video model from Kuaishou. It makes video with sound from words, a first frame, a last frame or up to 4 reference images, 3 to 15 seconds long.

Made by
Kuaishou
Makes
Video with sound
Starts from
Words, a first frame, a last frame or up to 4 reference images
Length
3 to 15 seconds
Shapes
16:9, 9:16, 1:1
Sound
Yes, when you turn it on
Prompt
Up to 2,500 characters
Price per run
$0.2772 to $1.848
On Leap since
October 5, 2026
Updated
October 5, 2026
  1. 01Describe it, or start from a photo
  2. 02See the exact price, then press go
  3. 03Keep it in your library, or take it further

Pick the right one

Kling O3 Reference and its siblings

ModelBest forLengthResolutionSoundPrice per run
Kling O3 Reference (this page)A video that keeps the people, products or style of up to four reference images, named in the prompt as @Image1 to @Image4, with an optional first and last frame and sound.3 to 15 seconds–Optional$0.2772 to $1.848
Kling 3.0Cinematic motion and strong physics, from text or a first frame.5 or 10 seconds––$0.924 to $1.848
Kling 2.6Kling's value model: smooth motion at a low price.5 or 10 seconds––$0.231 to $0.462
Kling 3.0 Motion ControlMoves the character in your image the way the person in a reference video moves, and keeps the face through turns, hands and hats. Billed by the length of the reference video.–Up to 1080p–$0.1386 a second of your video

Start with an idea

Bring a photo to Kling O3 Reference

Kling O3 Reference works on a picture you already have. Open it in the studio, add your photo, and see the result in seconds.

No prompt needed

Or start from a recipe

Recipes write the prompt for you: add a photo, pick a look, and get a finished result.

Why Leap

Why creators make it on Leap

  1. 01

    Every top model, side by side

    170 models from Black Forest Labs, ByteDance, Google, Kuaishou, Alibaba and more, side by side in one studio.

    See the models
  2. 02

    One click, or full control

    Character Studio, effects and recipes write the prompt for you. Or write your own for any model.

    See the tools
  3. 03

    Your work stays yours

    Everything you make stays in your library while your account is active, and you own what you make here.

    See what people make
  4. 04

    One balance for everything

    The studio, the API and your agents spend the same balance, and every run shows its price before it starts.

    How pricing works

Questions

Kling O3 Reference, answered

What is Kling O3 Reference?
Kling O3 Reference is a video model from Kuaishou. A video that keeps the people, products or style of up to four reference images, named in the prompt as @Image1 to @Image4, with an optional first and last frame and sound. On Leap it runs in one studio beside every other top model, with one account.
Can Kling O3 Reference start from a photo?
Yes. Kling O3 Reference starts from words, a first frame, a last frame or up to 4 reference images. The photo becomes the first frame, cropped to the video's shape.
How long can the videos be?
3 to 15 seconds. You choose the length before you press go, and the price follows it.
What resolution and sizes does it make?
16:9, 9:16 or 1:1.
Does it make sound?
Yes, when you turn sound on. Leave it off for a silent clip.
How is it different from Kling 3.0, Kling 2.6 and Kling 3.0 Motion Control?
Kling O3 Reference: A video that keeps the people, products or style of up to four reference images, named in the prompt as @Image1 to @Image4, with an optional first and last frame and sound. Kling 3.0: Cinematic motion and strong physics, from text or a first frame. Kling 2.6: Kling's value model: smooth motion at a low price. Kling 3.0 Motion Control: Moves the character in your image the way the person in a reference video moves, and keeps the face through turns, hands and hats. Billed by the length of the reference video. They all run in the same studio, so you can try one prompt on each.
How much does a run cost?
A 5-second clip without sound costs $0.462 at the default settings. By length and sound, a run costs $0.2772 to $1.848. You see the exact price before you press go, and a run that fails costs nothing.
Is Kling O3 Reference free to try?
Yes. Sign up with a Gmail, Outlook or work address and you start with $2 of free credit, no card needed. The cheapest run is $0.2772 (a 3-second clip without sound), so the credit covers seven of those. After that you pay per run, in dollars, with no subscription.
Can I use what I make commercially?
Yes. You own what you make on Leap, so you can use it commercially, and we never train AI on it. The people and brands shown in a result can still have rights in it; the Terms say more.
Is there an API?
Yes. Leap's REST API runs it as kling/kling-o3-reference, at the same price per run as the studio, and agents can call it through Leap's MCP server. The request is at the bottom of this page.
Do I need a Kuaishou account?
No. Leap sends your request to Kuaishou's model and gives you what it makes. One Leap account runs Kling O3 Reference and every other model, in the studio, through the API and from your agents.

Your idea, Kling O3 Reference.

Start free, no card. Every run shows its price before it starts.

Make it with Kling O3 Reference

For developers

Run Kling O3 Reference from code

Model ID
kling/kling-o3-reference
Makes
Video
Pricing
Per run, quoted first: POST /v1/quotes

Inputs

What the model takes. Leave out anything with a default.

Promptprompt
Text, up to 2,500 characters, required
First framefirst_frame
An imageAn image the video starts from. It is cropped to the video's shape.
Imagesimages
Up to 4 images, requiredUploaded files to edit or combine, by ID.
Last framelast_frame
An image
Aspect ratioaspect_ratio
16:9, 9:16, 1:1. Default: 16:9
Durationduration
3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 s, required. Default: 5 s
Soundaudio
On or off. Default: offGenerate dialogue, effects and music with the video.
Shot Typesettings.shot_type
Customize, Intelligent. Default: CustomizeThe type of multi-shot video generation.

The request

The same envelope for every model; only the model and its input change. With prefer: wait=60 the answer holds the result when the run finishes within a minute; a longer one is read again by its ID. A file input takes the ID that POST /v1/files returns for your upload. Agents can call it through the MCP server at app.tryleap.ai/api/mcp.

curl https://api.tryleap.ai/v1/generations \  -H "x-api-key: $LEAP_API_KEY" \  -H "content-type: application/json" \  -H "prefer: wait=60" \  -d '{    "model": "kling/kling-o3-reference",    "input": {      "prompt": "A paper boat drifting down a rainy street at dusk",      "images": [        "file_..."      ]    }  }' # Video often outlasts the wait and answers "queued" or "running".# Put its "id" in GENERATION_ID and read it again until it is done:# curl https://api.tryleap.ai/v1/generations/$GENERATION_ID -H "x-api-key: $LEAP_API_KEY"