Kling 4.0 Flash API

The faster, lower-cost Kling 4.0: 3 to 20 second clips at 720p with native audio from text, a first frame, reference images and videos, or a video to edit.

Kling AIVideo GenerationReleased Nov 4, 2026Proprietary EndpointNew

About Kling 4.0 Flash

The faster, lower-cost Kling 4.0: 3 to 20 second clips at 720p with native audio from text, a first frame, reference images and videos, or a video to edit.

Also known as Kling Flash, Kling AI Kling 4.0 Flash

video generationtext to videoimage to videoaudioediting

Kling 4.0 Flash specs

Model ID
kling-4-0-flash
Author
Kling AI
Category
Video Generation
Released
Nov 4, 2026
Input
TextImageVideo
Output
Video
Endpoints
POST/v1/videos/generations
Alternate model IDs
kling-4.0-flashkling/kling-4-0-flash

Kling 4.0 Flash API pricing

Live pay-as-you-go rates from the EmpirioLabs catalog. You are billed only for what you use, with no monthly minimum.

Type
Spec
Rate
Standard 720p
per second
$0.094
Video input 720p
per second of input plus output, at least twice the output
$0.063
Compare on the full pricing page

How to call the Kling 4.0 Flash API

Kling 4.0 Flash runs through POST /v1/videos/generations. The request returns a job_id right away; poll GET /v1/jobs/{job_id} until the job completes and read the output URLs from the result. Get an API key from the EmpirioLabs dashboard.

cURL: submit the job
curl https://api.empiriolabs.ai/v1/videos/generations \
  -H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-4-0-flash",
    "prompt": "Describe what you want Kling 4.0 Flash to generate."
  }'
cURL: poll for the result
curl https://api.empiriolabs.ai/v1/jobs/JOB_ID \
  -H "Authorization: Bearer $EMPIRIOLABS_API_KEY"
Python
import requests

response = requests.post(
    "https://api.empiriolabs.ai/v1/videos/generations",
    headers={"Authorization": "Bearer YOUR_EMPIRIOLABS_API_KEY"},
    json={
        "model": "kling-4-0-flash",
        "prompt": "Describe what you want Kling 4.0 Flash to generate.",
    },
)
job = response.json()

# Generation runs as an async job. Poll until it completes.
import time
while True:
    status = requests.get(
        f"https://api.empiriolabs.ai/v1/jobs/{job['job_id']}",
        headers={"Authorization": "Bearer YOUR_EMPIRIOLABS_API_KEY"},
    ).json()
    if status.get("status") in ("completed", "failed"):
        print(status)
        break
    time.sleep(5)
Full Kling 4.0 Flash API reference

Kling 4.0 Flash API parameters

Request parameters supported by the Kling 4.0 Flash API on EmpirioLabs. Defaults apply when a field is omitted.

ParameterTypeDefaultRange / valuesDescription
promptstring--Scene description, up to 26,000 English or 11,000 non-English characters. Refer to reference images as @image_1, @image_2 and videos as @video_1 in the order they...
resolutionenum720p720pOutput resolution.
aspect_ratioenum16:916:9, 9:16, 1:1, 21:9Output aspect ratio for text-to-video and references. A first frame on its own, or a video to edit, sets the aspect ratio instead.
durationnumber53 to 20Clip length in seconds, whole numbers only. A video edit keeps the length of the video it edits.
imagestring--First frame image URL. The video starts on this image.
reference_imagesstring--Reference image URLs for characters, objects, scenes or style, as a list or comma-separated. Up to 10 images in total.
videostring--Video URL to edit, 3 to 20 seconds. The result keeps its length, and bills at the video input rate.
reference_videosstring--Reference video URLs for motion, camera work or style, 3 to 20 seconds each and 20 seconds of video in total, as a list or comma-separated. Up to 5 videos in total....

Good to know

Generates 3 to 20 second clips at 24 fps in 720p, each with a native stereo audio track.

Inputs

  • Text to video: prompt only.
  • First frame (image): the video starts on this image and follows its aspect ratio.
  • Reference images (reference_images): characters, objects, scenes or style to use. A first frame can be sent with them.
  • Reference videos (reference_videos): motion, camera work or style to follow, 3 to 20 seconds each.
  • Video edit (video): a video to edit, 3 to 20 seconds. The result keeps its length.
  • Up to 10 images and 5 videos per request, with at most 20 seconds of video in total. In the prompt, refer to reference images as @image_1, @image_2 and to videos as @video_1, @video_2 (the video to edit first).
  • Last and middle frames and 1080p are available on Kling 4.0.

In the Playground

One attached image is the first frame, and two or more are references. The first attached video is the one edited; any others are references.

Media limits

  • Images: JPG or PNG, at least 300 px, aspect ratio between 2:5 and 5:2, up to 30 MB.
  • Videos: MP4 or MOV, 3 to 20 seconds, 700 to 4553 px per side, 24 to 60 fps, up to 300 MB.

Billing

  • Without a video input: billed per second of the generated clip.
  • With a video input (a video to edit or a reference video): billed at the video input rate on twice the length of the generated clip, or on the input plus the output when the input is longer.

Kling 4.0 Flash API: common questions

How much does the Kling 4.0 Flash API cost?

On EmpirioLabs, Kling 4.0 Flash is billed pay as you go: Standard 720p $0.094 per second; Video input 720p $0.063 per second of input plus output, at least twice the output. The live rate card on this page always matches what the API charges.

Which endpoint does Kling 4.0 Flash use?

Kling 4.0 Flash is served through POST /v1/videos/generations on api.empiriolabs.ai with standard bearer-token authentication.

Can I try Kling 4.0 Flash in the browser before integrating?

Yes. The EmpirioLabs playground runs Kling 4.0 Flash in the browser with the same parameters the API exposes, so you can test prompts before writing code.

How do I get a Kling 4.0 Flash API key?

Create an EmpirioLabs account, then generate a key under API Keys in the dashboard. Billing is pay-as-you-go credits, so you only pay for the requests you make.

Ready to use better endpoints?

Check out our pricing or reach out if you want your own model deployed on our stack.