Kling Image 4.0 is the upcoming image model from Kling AI. Kling has announced it as coming soon but has not published its specs, pricing or release date. The workspace below follows the official Kling Image 3.0 Omni parameters until Kling Image 4.0 details are available.
Kling Image 4.0 API workspace
Kling has not published Kling Image 4.0 specs yet. This workspace follows the official Kling Image 3.0 Omni parameters, and RUN stays disabled until launch.
Up to 2,500 characters. Refer to reference images by their order, such as <<<image_1>>>.
Up to 10 public http(s) image URLs. Kling Image 3.0 Omni accepts JPG or PNG up to 10 MB, at least 300 px, with an aspect ratio between 1:2.5 and 2.5:1.
Your images will appear here
Kling AI describes Kling Image 4.0 as "stunningly realistic, sharp, precise" and says it is coming soon. Until Kling publishes the details, this page sticks to what is official, and the workspace lets you prepare kling-image-4-0 requests using the parameters of the current Kling Image 3.0 Omni model.
These Kling Image 4.0 API examples use APIMart's common asynchronous task format and the workspace parameters, which follow Kling Image 3.0 Omni until Kling publishes the Kling Image 4.0 specs.
Save your APIMart API key as APIMART_API_KEY. The Kling Image 4.0 API examples send Authorization: Bearer <key> for submission and task queries.
POST /v1/images/generations with model set to kling-image-4-0, your prompt, size and resolution. Add image_urls for reference images. Read the task ID from data[0].task_id.
Use GET /v1/tasks/{task_id}. For a completed task, read data.result.images[0].url[0]; for a failed task, read data.error.message.
See the APIMart documentation for APIMart authentication and general API usage.
curl -X POST https://api.apimart.ai/v1/images/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kling-image-4-0",
"prompt": "A rain-soaked neon street at night, a cyclist in a yellow raincoat passing a ramen stall, cinematic lighting and sharp reflections",
"size": "16:9",
"resolution": "2k",
"n": 1
}'import os
import time
import requests
API_BASE = "https://api.apimart.ai"
headers = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}"}
# 1. Submit the generation task
payload = {
"model": "kling-image-4-0",
"prompt": "A rain-soaked neon street at night, a cyclist in a yellow raincoat passing a ramen stall, cinematic lighting and sharp reflections",
"size": "16:9",
"resolution": "2k",
"n": 1
}
resp = requests.post(f"{API_BASE}/v1/images/generations", headers=headers, json=payload, timeout=60)
resp.raise_for_status()
task_id = resp.json()["data"][0]["task_id"]
# 2. Poll the task until it finishes
for _ in range(200):
task_resp = requests.get(f"{API_BASE}/v1/tasks/{task_id}", headers=headers, timeout=30)
task_resp.raise_for_status()
task = task_resp.json()["data"]
if task["status"] == "completed":
print(task["result"]["images"][0]["url"][0])
break
if task["status"] == "failed":
raise RuntimeError(task.get("error", {}).get("message", "generation failed"))
time.sleep(3)
else:
raise TimeoutError(f"task {task_id} did not finish in time; continue GET polling with this task ID")curl -X POST https://api.apimart.ai/v1/images/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kling-image-4-0",
"prompt": "Place the sneaker from <<<image_1>>> on a wet concrete step at golden hour, low angle, shallow depth of field",
"image_urls": [
"https://example.com/sneaker.png"
],
"size": "1:1",
"resolution": "2k",
"n": 1
}'TASK_ID="task_xxx" # replace with the task_id returned by the POST request
curl "https://api.apimart.ai/v1/tasks/$TASK_ID" \
-H "Authorization: Bearer $APIMART_API_KEY"| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
model | string | Required | Default: kling-image-4-0 | The model value used in this Kling Image 4.0 workspace and its examples. The final model ID will be confirmed when the model goes live. |
prompt | string | Required | Default: — | Image description, up to 2,500 characters. Refer to reference images as <<<image_1>>>, <<<image_2>>> and so on, in array order. |
image_urls | string[] | Optional | Default: — | Up to 10 reference images as public http(s) URLs. Kling Image 3.0 Omni accepts JPG or PNG files up to 10 MB, at least 300 px, with an aspect ratio between 1:2.5 and 2.5:1. |
size | string | Optional | Default: auto | Aspect ratio: 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9 or auto. |
resolution | string | Optional | Default: 1k | 1k, 2k or 4k. |
result_type | string | Optional | Default: single | single for separate images, series for a set of related images. |
n | integer | Optional | Default: 1 | Number of images, 1 to 9. Only used when result_type is single. |
series_amount | integer | string | Optional | Default: 4 | Number of images in a series, 2 to 9, or auto to let the model decide. Only used when result_type is series. |
watermark | boolean | Optional | Default: false | Whether to also return a watermarked result. |
What Kling has confirmed about Kling Image 4.0, and what you can prepare in the APIMart workspace today.
Get your requests ready now and update them when Kling publishes the Kling Image 4.0 specs.
Create your API key, set up the prompt and reference images, then review the JSON.
Create an API key in the APIMart console. The same key works for every model on APIMart, and requests send it as Authorization: Bearer <key>.
Describe the image, add up to 10 reference images as http(s) URLs, then choose single images or a series, the aspect ratio and the resolution.
Switch to JSON to check every field. RUN stays disabled until Kling Image 4.0 is available on APIMart.
Answers about Kling Image 4.0 availability, the workspace parameters, API keys and naming.
Kling Image 4.0 is the next image generation model from Kling AI, Kuaishou's AI creative platform and the team behind the Kling video models. Kling has announced it as coming soon and describes it as "stunningly realistic, sharp, precise".
Not yet. This page is a preview: you can prepare and check kling-image-4-0 requests in the workspace, but RUN stays disabled and no tasks are submitted until the model is available on APIMart.
Kling has not announced a release date. Its official announcement only says that Kling Image 4.0 is coming soon. This page will be updated when Kling publishes more details.
Kling has not published specs for Kling Image 4.0 yet, including resolutions, aspect ratios, reference limits or prompt length. Specs you may see on other sites are not confirmed by Kling.
Kling Image 3.0 Omni is the most capable model in the Kling Image API lineup with published documentation. Using its parameters lets you prepare realistic requests now. We will check every parameter against the official Kling Image 4.0 documentation when it is released.
The workspace and API examples use kling-image-4-0. APIMart will confirm the final model ID when Kling Image 4.0 goes live, and this page will be updated if it changes.
Kling has not published pricing for Kling Image 4.0, and APIMart pricing is not set yet. Prices will appear on this page when the model launches on APIMart.
Create an API key in the APIMart console. One APIMart API key works for all models, so you will not need a separate key for Kling Image 4.0.
Up to 10, following Kling Image 3.0 Omni. Enter them as public http(s) URLs and refer to them in the prompt as <<<image_1>>>, <<<image_2>>> and so on. Kling Image 3.0 Omni accepts JPG or PNG files up to 10 MB.
1K, 2K or 4K, and 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9 or auto. These match Kling Image 3.0 Omni; Kling has not yet confirmed the options for Kling Image 4.0.
With result_type set to single, you get 1 to 9 separate images (n). With series, you get a set of 2 to 9 related images that keep the same characters and style (series_amount), or auto to let the model decide.
No. Kling 4.0 and Kling 4.0 Flash are Kling's video models and are already available on APIMart. Kling Image 4.0 is a separate image model that has been announced but not released.
Kling has not published a comparison yet. For reference, Kling Image 3.0 Omni already offers native 2K and 4K output and series generation. We will add the official differences when Kling releases them.
Yes. Kling sometimes styles the name as "Kling IMAGE", and Kling AI Image 4.0 simply adds the company name. You may also see Kling Image 4, Kling 4.0 Image, Kling-Image-4.0 or KlingImage 4.0. They all refer to the same upcoming model, which this workspace calls kling-image-4-0.
Explore more models in the same category.

MiniMax H3 Image
Built for image quality and prompt adherence. It handles text-to-image, reference-based generation and image editing. MiniMax says it can generate a 4K (4096 × 4096) image in under 10 seconds and work with up to 16 reference images. In Chinese launch coverage, MiniMax puts its speed at about 1.5 to 2 times that of mainstream tools.

MiniMax H3 Image Highspeed
Built for speed. The same coverage puts it at about 3 to 4 times faster than mainstream tools, which suits batch jobs or anything that needs fast turnaround.

Seedream 5.5
Seedream 5.5 brings your imagination to life through expressive visuals. From dreamy illustrations to cinematic scenes and product concepts, explore color, texture, and light to shape images that tell your story.

GPT Image 2.6 Sunburst
GPT Image 2.6 Sunburst brings a fresh burst of creativity to your ideas. From vibrant illustrations to atmospheric scenes and striking product concepts, explore rich colors, expressive lighting, and thoughtful compositions that make your vision shine.