About Kling O3
Kling O3 video edit takes an existing clip in video_url and a prompt describing only the change, and returns the same shot with that change applied.
Call it with the model name kling/o3/video-edit on POST /jobs/createTask, from 18.78 credits per second ($0.19). New accounts get 30 free credits to try it in the Playground first.
Introduction
Kling O3 Video edit changes a video you already have instead of generating one from nothing. Upload the source clip and write a prompt describing only what should change - season, wardrobe, backdrop, style - and everything you leave out stays as it was filmed. Optionally add up to 4 reference photos to bring in a new character, outfit or object. The result runs exactly as long as your source clip, from 3 to 10 seconds, at 720p or 1080p.
Choose a Kling O3 mode by starting material
| What you have at the start | Mode |
|---|---|
| a written scene and nothing else | Kling O3 Text to video |
| an approved opening frame | Kling O3 Image to video |
| up to 7 reference photos | Kling O3 Reference to video |
| a video you already have | Kling O3 Video edit |
Key features
Change only what you name
The prompt is a list of changes, not a full scene description: say the season should turn to snow, or the wardrobe should switch to a winter coat, and the framing, motion and pacing of the source clip carry over untouched.
Length and shape follow your source
There is no duration field to set - the edited clip matches the length of the video you uploaded, from 3 to 10 seconds, and its resolution is limited to 720p or 1080p rather than the 4K available on the other three modes.
Reference photos for what the source never showed
Add up to 4 reference images when the edit needs to introduce something the source clip does not contain - a specific new outfit, prop or look - alongside the changes described in the prompt.
Request parameters
| Parameter | Required | Description |
|---|---|---|
elements | No | One named subject to bring into the edited clip, built from 2 to 4 images of the same character or object. Write @name in the prompt where it should appear. |
keep_original_audio | No | Keep the source clip's own sound in the result. Default: true |
prompt | Yes | Say what to change in the clip: the subject, outfit, setting, style or weather. Anything you do not mention stays as it is. |
reference_image_urls | No | Pictures of the people, outfits or places the result should keep. |
resolution | No | Output size. Bigger sizes look sharper and cost more. 720p · 1080p · Default: 720p |
video_url | Yes | The clip to work on. |
Pricing
Prices are in credits, the same credits you use on the website (1 credit = $0.01). The hold is taken when the task is accepted and settled to the actual charge when it finishes; failed tasks are refunded.
| Option | Credits per second | ≈ USD |
|---|---|---|
| resolution 720p | 18.78 | $0.19 |
| resolution 1080p | 25.2 | $0.25 |
For an exact number before you run, send the same body to POST /jobs/quote.
Use cases
Reuse a clip in a different season or weather
Take a shot you already like and change only the season or weather - summer to snow, clear sky to rain - without re-filming or regenerating the whole scene from scratch.
Swap wardrobe or backdrop on an approved take
When the framing and performance of a clip are right but the outfit or the location behind the subject needs to change, video-edit updates just that element.
Restyle a clip you already generated elsewhere
Bring in a video from a text-to-video or image-to-video run on this or another model and push it toward a different visual style using the prompt, optionally guided by reference photos.
Prompt guide & pro tips
Write it as a list of changes, not a scene
Start from what already exists in the clip and name the deltas: "change the season to snow," "put him in a charcoal coat instead." A full scene description invites the model to redo more than you intended.
Check your source video's dimensions first
The source needs to be roughly 720 pixels or more on its shortest side. A clip that is too small will be rejected before generation starts, so verify this ahead of a job that also includes reference photos.
Add references only for what the edit introduces
Reference photos are for bringing something new into the shot - an outfit, an object, a look the source clip does not have. For changes that are purely descriptive, like a season or a color shift, the prompt alone is usually enough.
Spell out distinct wardrobe per person
When a clip has more than one person and each needs a different new outfit, name them separately - "the rider in front in a navy coat, the one behind in a grey coat" - rather than one shared description. Left unspecified, our winter test gave both riders the same coat.
API quick start
Call the Kling O3 API in three steps
- Create an API key. Sign in to PopNow and create a key in the API console. Keys start with sk-popnow- and use the same credits as the website.
- Send the request. POST /jobs/createTask with model kling/o3/video-edit and an input object that follows the parameter table. The response returns a task ID straight away.
- 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: $IDEMPOTENCY_KEY" \
-d '{
"model": "kling/o3/video-edit",
"input": {
"prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
"video_url": "https://media.popnow.ai/effects/flower-bloom-1abd6efb.mp4",
"resolution": "720p"
}
}'
# 2. Copy data.taskId from the response, then poll until state is succeeded, failed, canceled or expired.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$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.
About the Kling O3 Edit Video endpoint
Add reference_image_urls when the change involves a specific object, such as a jacket, and point at it in the prompt.
Billing follows the source clip length at the edit rate, at 720p or 1080p only.
Kling O3 API questions
How much does the Kling O3 API cost?
Kling O3 starts at 18.78 credits ($0.19) per second with Edit Video, its cheapest task. Prices by option: 720p 18.78 credits per second, 1080p 25.2 credits per second. Credits are the same as on the website (1 credit = $0.01), new accounts get 30 free credits, and failed tasks are refunded.
What are the limits of the Kling O3 · Edit Video API?
Resolutions: 720p · 1080p; Reference images: 4; Prompt length (characters): 2,500; Largest upload: 10 MB. Requests outside these limits are rejected before anything is charged.
More about Kling O3
- Endpoints
- All Kling O3 endpoints and pricing
- Compare
- Kling 3.0 vs Kling O3