Seedream 5.5 is live on Atlas Cloud — native transparent PNG & 80+ round editing
Home
Explore
Kwaivgi
Kling 4.0
kwaivgi/kling-v4.0/video-edit
Atlas Cloud GeneratorUnlock your potential as a director.Go Create
Kling Video 4.0 Video-Edit
video-to-video

Kling Video 4.0 Video-Edit API by Kuaishou

kwaivgi/kling-v4.0/video-edit
Video-edit

Kling 4.0 video editing: restyle, replace or add content in an existing clip using text plus optional reference images, videos, subjects and voices.

Kling Video 4.0 Video-Edit is developed by Kuaishou. Atlas Cloud (operated by Atlas Cloud AI LLC) provides access to it and does not own it. All trademarks belong to their respective owners.

Kling Video 4.0 Video-Edit

Kling Video 4.0 Video-Edit makes localized edits and style changes to an existing clip from a text instruction — swap a background, change the weather, restyle the whole shot — optionally guided by reference images, videos, subjects and voices.

Why Choose This?

  • Precise local edits. Change what you ask for and keep the rest of the clip.
  • Style adjustments. Restyle the full video while keeping subject and camera motion.
  • Reference-guided. Bring in a product, a character or a look from references.

Parameters

ParameterRequiredDescription
modelYeskwaivgi/kling-v4.0/video-edit
promptYesEditing instruction. The video to edit is @video_1; other inputs are @image_N, @video_2…, @element_N, @voice_N.
videoYesVideo to edit. MP4 / MOV, max 300MB, 3–30s, 700–4553px.
reference_imagesNoReference images (URLs or Base64).
reference_videosNoFeature reference videos, up to 4.
elementsNoSubjects created for this request from images or a video clip (see Notes).
voice_idsNoKling voice IDs, up to 3.
durationNoOutput length in seconds, integer 3–30. Default 5.
resolutionNoNative 720p or 1080p (default); Tencent ESR 1080p-esr, 2k-esr, or 4k-esr.

How to Use

curl -X POST "https://api.atlascloud.ai/api/v1/model/generateVideo" \ -H "Authorization: Bearer $AIP_API_KEY" \ -H "Content-Type: application/json" \ --data-raw '{ "model": "kwaivgi/kling-v4.0/video-edit", "prompt": "Turn the scene into a snowy winter evening while keeping the subject and camera motion of @video_1 unchanged.", "video": "https://p2-kling.klingai.com/kcdn/cdn-kcdn112452/kling-qa-test/animals-video-5s.mp4", "duration": 5, "resolution": "1080p" }'

Generation is asynchronous. The response contains a request id; poll GET https://api.atlascloud.ai/api/v1/model/prediction/{id} until status is completed, then read the video URL from outputs.

Notes

  • Preview. Kling 4.0 is in early access upstream; pricing and limits follow Kling's preview terms and may change.
  • The Kling 4.0 API has no audio on/off or multi_shot parameter; describe sound and shot structure in the prompt.
  • duration must be an integer from 3 to 30. Supported resolutions are 720p, 1080p, 1080p-esr, 2k-esr, and 4k-esr.
  • ESR choices generate a Kling source video, then enhance it with Tencent ESR: 1080p-esr uses a native 720p source; 2k-esr and 4k-esr use a native 1080p source. These are enhanced outputs.
  • 2k-esr targets 1440p (2560×1440 at 16:9); 4k-esr targets 2160p (3840×2160 at 16:9). The source aspect ratio and frame rate are preserved; portrait outputs use the corresponding short-edge target.
  • ESR pricing adds an enhancement charge for output seconds to the source-generation price. Video-input billing still applies to source generation; the enhancement charge is not multiplied by input-video duration. ESR requires an additional processing stage, so completion can take longer than native generation.
  • The output video is transferred to AtlasCloud storage before it is returned.
  • Subjects (elements): describe each subject in the request with element_name (and an optional element_description) plus reference_type — image_refer: frontal_image + 1–3 refer_images; video_refer: one 3–8s refer_videos clip of a real person. The subjects are created first and the video is generated right after, in the same request; this adds a short wait before generation starts.
  • Subjects are created anew for every request and are not saved: no subject ID is returned, and there is no element_id to reuse. To keep a character consistent across generations, send the same subject images or clip again.
  • Limits per request: up to 10 images (keyframes + reference images), 5 videos (3–30s in total), 7 subjects (at most 3 video character subjects), 3 voices, and 15 items in total.
  • Set duration to the length you want back; it is billed as the output duration.

Explore Similar Models

One API for All Media AI.

Explore all models