Edit a video by describing the change. Takes up to three reference images to guide who or what appears, and can keep the original audio. Outputs up to 10 seconds.
wan-27-video-editThe price changes with the length of the audio or video you send and the resolution. The estimate call returns the exact figure for your input, free.
API keys come with the Scale and Ultra plans
The smallest request that works: the required fields only.
curl https://api.percify.io/v3/playground/v1/run \
-H "Authorization: Bearer $PERCIFY_API_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"modelId": "wan-27-video-edit",
"input": {
"video": "https://example.com/clip.mp4",
"prompt": "A red fox running through fresh snow at sunrise"
}
}'9 fields in input, required first.
videoRequiredThe source video to edit.
promptRequiredThe positive prompt for the generation.
seedOptionalThe random seed to use for the generation. -1 means a random seed will be used.
imagesOptionalList of reference images for video editing (0-3 images, optional).
durationOptionalDuration of the output video in seconds. 0 means use input video length (max 10s). Set 2-10 to trim from 0s to specified length.
resolutionOptionalThe resolution of the generated video.
audio_settingOptionalAudio setting. 'auto' (default): model decides based on prompt content. 'origin': keep original audio from input video.
negative_promptOptionalThe negative prompt for the generation.
enable_prompt_expansionOptionalIf set to true, the prompt optimizer will be enabled.

Pick a use case, run it in the playground, copy the call.
API keys come with the Scale and Ultra plans