by Kuaishou

Two to ten matching images from one brief, with the same character, product and palette

kling/image-4.0-set/text-to-image
  • Up to 8K
  • Up to 10 reference images
  • 8 aspect ratios
  • Up to 10 per run
  • Create Image

About Kling Image 4.0 Set

Two to ten matching images from one brief, with the same character, product and palette.

Call it with the model name kling/image-4.0-set/text-to-image on POST /jobs/createTask, from $0.063.

Introduction

Build a sequence around a shared subject

The Kling Image 4.0 Set endpoint accepts a brief for two to ten related pictures. Describe the series as a whole, then give each image a position and a visual purpose. Optional reference images apply across the request, so a product picture can anchor a lighting study or a set of viewing angles.

The distinction to plan around is continuity: decide which details should remain recognizable and which should change. For a lighting study, name the object and its placement as fixed details, then assign a different light source to each image. For a process sequence, define the workspace once and describe the task at each stage.

Request parameters

These are the public fields for this endpoint. Keep the requested image count aligned with the numbered descriptions in your prompt.

FieldRequiredChoices and limitsRole
promptYes1–21,000 charactersDescribe the series, the images in order, and the details to preserve. Cite references as @image_1, @image_2.
reference_image_urlsNoUp to 10 JPG or PNG filesShared visual references. Each file may be up to 30 MB, must measure at least 300 px per side, and must have a ratio between 2:5 and 5:2.
num_outputsNo2–10; default 4Number of related images requested.
resolutionNo2k, 4k, 8k; default 2kResolution tier used to determine the image rate.
aspect_ratioNoauto, 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9; default autoShape setting for the series; auto chooses from the prompt and references.

Count the images and the references separately

A reference supplies visual input; it does not set the number of pictures in the series. The rose request used one reference with num_outputs set to two. Select the count deliberately rather than assuming that the number of uploads determines it.

Budget for the delivered series

Review the output count, resolution, and reference count together before submitting your request.

Write a brief with fixed details and planned changes

Begin with the shared subject, setting, and visual treatment. Follow with numbered image descriptions and a clear preservation instruction. Spend the prompt space on details a reviewer can check, such as product proportions, shelf placement, camera view, and lighting.

Assign one visual job per image

For a product series, one image might establish the object and another might examine it under different light. For a workshop process, each image can depict a named stage. Describe what distinguishes each picture so the request has a useful progression.

Give references a named role

Identify the product in @image_1 or the setting in @image_2, then say what the series should retain from each. Make sure those numbers match the input order.

Keep shape and resolution deliberate

The series has a shared aspect ratio setting and a shared resolution tier. If a square layout is required, choose 1:1 rather than leaving the shape to auto. The rose run selected 1:1 and 2k. Review the composition and the retained subject details before deciding whether a different tier is appropriate for your work.

What the rose request returned

On November 18, 2026, we submitted a prompt asking for the referenced product on a wooden shelf in two lighting conditions. The settings were one reference, two outputs, 2k resolution, and a 1:1 aspect ratio. The request returned two 2048 × 2048 PNGs.

Morning and evening from the same brief

The morning image shows the rose on the wooden shelf in pale light. Its companion shows the same rose on the same shelf under a warm table lamp. Both views came from the same request. Compare the two images together when judging the preservation goal and the requested change in illumination.

Review the group before selecting images

Place the returned pictures together and compare them against the shared requirements in your prompt. For a product study, inspect the object's proportions and position as well as the planned lighting changes. For a process board, check that every requested stage is present and that the workspace details connect the sequence. If a view needs revision, make its image description more specific while keeping the shared instructions clear.

Limits and related choices

The image count must remain between two and ten, and the prompt must fit within 21,000 characters. Check every reference against the file, dimension, and ratio requirements in the table. Series generation still needs a visual review: compare product details and the intended changes across all returned pictures before using them in a catalog or campaign.

For a standalone image brief, compare Kling Image 4.0. For a continuous page composition, consider Kling Image 4.0 Long: Kling says its long-image mode produces one complete page image. Choose according to whether your deliverable needs separate images or a continuous page layout.

Request parameters

ParameterRequiredDescription
aspect_ratioNoThe shape of the frame: wide, square or tall. auto · 16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 3:2 · 2:3 · 21:9 · Default: auto
num_outputsNoHow many results to make in one go. ≥ 2 · ≤ 10 · Default: 4
promptYesDescribe what you want to make.
reference_image_urlsNoPictures of the people, outfits or places the result should keep.
resolutionNoChoose an available output size. 2k · 4k · 8k · Default: 2k

Pricing

Prices are in USD and charged to your PopNow balance. When a task is accepted, we reserve the maximum quoted amount from your API balance. On completion, we charge for actual usage, never more than the hold, and release the remainder. If the task fails, the hold is released.

OptionUSD per image
resolution 2K$0.063
resolution 4K$0.094
resolution 8K$0.157

Send the same request body to POST /jobs/quote to see the estimated and maximum USD charge before running. A quote does not run the task or charge your balance.

API quick start

Call the Kling Image 4.0 Set API in three steps

  1. Create an API key. Sign in to PopNow and create a key in the API console. Keys start with sk-popnow-. API calls are billed in USD from your PopNow balance.
  2. Send the request. POST /jobs/createTask with model kling/image-4.0-set/text-to-image and an input object that follows the parameter table. The response returns a task ID straight away.
  3. Get the result. Poll GET /jobs/recordInfo with the task ID, or pass a callBackUrl and we post the result to it. Download the output links when the task succeeds.
# 1. Create the task. Keep the Idempotency-Key if you retry after a network error.
export IDEMPOTENCY_KEY="$(uuidgen)"
curl -X POST https://api.popnow.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $POPNOW_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $IDEMPOTENCY_KEY" \
  -d '{
  "model": "kling/image-4.0-set/text-to-image",
  "input": {
    "prompt": "A ceramic teapot on a linen cloth, soft window light, muted palette",
    "num_outputs": 4,
    "resolution": "2k",
    "aspect_ratio": "auto"
  }
}'

# 2. Copy data.taskId from the response, then poll until state is succeeded, failed, canceled or expired.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $POPNOW_API_KEY"

Replace the prompt and any links with your own. The API tab has the response, polling and callback reference.

Kling Image 4.0 Set API questions

How much does the Kling Image 4.0 Set API cost?

Kling Image 4.0 Set costs $0.063 per image. Prices by option: 2K $0.063 per image, 4K $0.094 per image, 8K $0.157 per image. API usage is charged in USD to your PopNow balance. When a task is accepted, we reserve the maximum quoted amount from your API balance. On completion, we charge for actual usage, never more than the hold, and release the remainder. If the task fails, the hold is released.

What is the model name for Kling Image 4.0 Set in the API?

Put kling/image-4.0-set/text-to-image for Create Image in the model field of POST /jobs/createTask.

How is a set different from generating several images?

A set is briefed as a related sequence. Describe the shared subject and the role of each image, then inspect the returned group for continuity rather than treating every picture as a separate concept.

Do I need reference images for a series?

No. You can submit a written brief without references. Add them when an existing product or other visual detail matters, and cite each source by its image number.

Can every image in a series use its own aspect ratio?

No. The request has one aspect ratio setting and one resolution setting for the series. Choose the shape you need, or use auto to let the model choose from the prompt and references.

How many images should I request?

Request between two and ten images. The default is four; match the count to the number of image descriptions in your brief so each output has a defined purpose.

What did the product-reference run show?

It returned two 2048 × 2048 PNGs. The same rose appears on the same wooden shelf, with pale morning light in one image and warm table-lamp light in the other.

What are the limits of the Kling Image 4.0 Set API?

Resolutions: 2K · 4K · 8K; Aspect ratios: auto · 16:9 · 9:16 · 1:1 · 4:3 · 3:4 +3; Outputs per call: 2–10; Reference images: 10; Prompt length (characters): 21,000. Requests outside these limits are rejected before anything is charged.

Can I try Kling Image 4.0 Set without writing code?

Run Kling Image 4.0 Set in the Playground with your API USD balance, without an API key. Each run appears in your developer history with its matching code.

All Kling models

More about Kling Image 4.0 Set