Skip to main content

Kling Image 4.0 Set, from Kuaishou's Kling team, generates two to ten coordinated images from one brief, with shared characters, products, and palette across a storyboard or campaign at up to native 8K.

From $0.063/image

Workbench

Input

Text to image
159/21000

Reordering updates prompt references. Removing media also removes its references; review your prompt.

Required, up to 21,000 characters. Give the series-wide rules, then describe each image in order. Cite any uploads as @image_1, @image_2, and so on. at least 1 characters · at most 21000 characters
affects price1/10

Drop files or media links, paste, or browse · Up to 10 files · Max 10 MB · JPG · PNG

  • @image_1
1 / 10 files
Optional, up to 10 images shared by the entire series. Files must be JPG or PNG, 30 MB at most, 300 px or more per side, with a ratio from 2:5 to 5:2. Each one adds an input charge. at most 10 items
affects price
Series tier: 2k, 4k, or 8k. The default is 2k, and larger tiers raise the price of every image.
Shape used by the whole series: auto, 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, or 21:9. The default is auto, and the choice has no effect on the charge.
Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run
rose-shelf-morning-eveningExample result

Missing the model you need?

KLING · IMAGE 4.0 SET

Kling Image 4.0 Set API at a glance

Two to ten coordinated images from one brief.

Linked images
2–10
Every picture in the series is planned together and returned in sequence.
Shared references
10
Characters, outfits, products, and style images apply across the whole series.
Highest tier
8K
The series runs at one resolution: 2k, 4k, or 8k.

Results

What you can make with Kling Image 4.0 Set

  • A rose on a wooden shelf in pale morning light.

    Observed product continuity

    A rose on the same shelf in two lights

    Our two-image request used one product reference and asked for the product to stay identical. The morning output shows the same rose on the same wooden shelf as the evening view. Both returned files were 2048 × 2048 PNGs at the 2k setting.

    Morning output from the two-image rose request; 2048 × 2048 PNG at 2k.

  • The same rose on the same wooden shelf under a warm table lamp.

    A distinct evening frame

    Warm lamp light changes the scene

    The companion output puts the rose and wooden shelf under a warm table lamp. It came from the same request as the pale morning image, with two outputs and a square canvas. Compare the pair to see how the requested lighting change sits alongside the retained subject and shelf.

    Evening output from the same two-image request; 2048 × 2048 PNG at 2k.

  • Series controls

    Give every frame a defined role

    Write the shared requirements once, then describe the images in sequence. The request accepts two to ten outputs, up to ten shared references, and a resolution choice of 2k, 4k, or 8k. Count, shape, and resolution are separate decisions; make each agree with the series you intend to review.

Before you run it

What to know before you run Kling Image 4.0 Set

All modes
A series holds at least two images and at most ten. Picking the count automatically is not offered, so the price is known before the run starts.
All modes
Optional references follow the same file rules throughout: JPG or PNG, 30 MB at most, 300 px or more per side, ratio from 2:5 to 5:2.

Use cases

What to build with the Kling Image 4.0 Set API

  • Product lighting studies

    Compare a product under different lighting conditions while asking for the object and its placement to remain fixed. Our rose request returned a morning view and an evening lamp view of the same shelf scene.

  • Craft process boards

    Describe a workshop task as numbered stages and match num_outputs to the planned sequence. Name the tools, materials, and workspace details to carry through the process board.

  • Catalog angle planning

    Assign each image a different product angle, with a shared reference and a written list of details to preserve. Inspect the returned objects against that reference before selecting catalog material.

  • Campaign composition studies

    Use a common subject and palette across a series, giving each image a distinct composition. Choose a supported ratio for the intended placement and review the group for visual continuity.

Prompts

Kling Image 4.0 Set prompt examples you can copy

  • Text to image

    Rose shelf lighting pair

    A two-image set showing the product in @image_1 on a wooden shelf: image 1 in morning light, image 2 in evening lamp light. Keep the product identical in both.

    Why it works. The reference identifies the product, the image numbers assign different lighting conditions, and the final instruction names what should stay identical.

    Try this prompt
  • Text to image

    Ceramic bowl viewing angles

    Create a three-image product series using the ceramic bowl in @image_1. Image 1: front view at shelf height. Image 2: view from above showing the interior. Image 3: side view showing the foot ring. Keep the glaze color, proportions, and plain gray studio setting consistent.

    Why it works. Each view examines a specific part of the bowl, while the glaze, proportions, and studio setting give the series a shared visual anchor.

    Try this prompt
  • Text to image

    A bookbinding process board

    Create a four-image sequence in one bookbinding workshop. Image 1: folded paper sections on a workbench. Image 2: a professional bookbinder in a navy work shirt stitching the sections. Image 3: the stitched block beside its cloth cover. Image 4: the completed book resting closed. Keep the bench, tools, blue cover cloth, and soft window light consistent.

    Why it works. Named stages establish the process order, while the bench, tools, cover cloth, and window light connect the four images.

    Try this prompt
  • Text to image

    Office lobby material studies

    Make a three-image series of one office lobby with an oak reception desk, pale stone floor, and curved white wall. Image 1: wide view from the entrance. Image 2: closer view of the desk and wall junction. Image 3: view toward the seating area. Keep the layout, furniture, materials, and daylight direction consistent.

    Why it works. Wide and close views give the lobby study a clear progression, with the furniture, materials, and daylight direction tying the compositions together.

    Try this prompt
  • Text to image

    Professional studio portrait views

    Create a three-image studio portrait series of a fictional architect in her forties wearing a charcoal blazer and white shirt. Image 1: front-facing head-and-shoulders portrait. Image 2: three-quarter view. Image 3: profile view. Maintain the same facial features, hairstyle, clothing, neutral background, and soft side light.

    Why it works. Three named viewing angles make the portrait sequence easy to plan, while the clothing, background, and light provide concrete details to preserve.

    Try this prompt

Pricing

Kling Image 4.0 Set API pricing

One image starts at $0.063 at 2K and $0.157 at 8K.

Get started

How to use the Kling Image 4.0 Set API in three steps

  1. Step 1

    Define the shared subject

    Describe the product, place, or professional subject that connects the series. State the details you want preserved before listing the individual pictures.

  2. Step 2

    Assign a purpose to each image

    Number the images in the order you need them and describe one view or moment per image. Set num_outputs to that count, between two and ten.

  3. Step 3

    Map references and set the canvas

    Add shared references when you need a visual anchor, citing them as @image_1 and @image_2. Select the resolution and aspect ratio for the series.

Parameter reference and code samples are in the API and Readme tabs at the top of this page.

FAQ

Kling Image 4.0 Set API FAQ

About Kling Image 4.0 Set

Plan the relationship between images before choosing the output count. A product study might keep the object and shelf fixed while changing the light; a process board might keep the workspace fixed while the task advances. Our rose request returned two square PNGs at 2k, showing the same shelf scene in morning and evening light. Use that pair as a concrete reference for your review, then judge every image in your own set against its brief. Check the current resolution rate and reference charges, and save the prompt with the returned sequence for later revisions.

Related models and next tasks