Alibaba Wan - video edit (3.0)

Alibaba's Wan 3.0 edits a video from a plain instruction: swap objects, restyle, relight or change the dialogue, and keep everything else as it was.

Alibaba Wan 3.0 edits the video you give it from a plain-language instruction: add, remove or replace an object or a person's clothes, turn the whole scene into another style such as clay or anime, change the lighting or the time of day, or rewrite what a speaker says. Whatever the prompt does not mention stays as it was: the motion, the other objects and the sound. Add reference images to show what to put in (the prompt calls them "Image 1", "Image 2") or reference audio for the voice.

Inputs: prompt (required: what to change; "Edit the video:" is added when the prompt does not say it), video (required; MP4 or MOV, 1-15 s, up to 100 MB), reference_images (up to 10), reference_audios (up to 5), resolution (480P, 720P by default, 1080P), ratio (adaptive by default: the result keeps the video's shape), duration (-1 by default: the result keeps the video's length; or 2-30 seconds in all), audio, prompt_extend, watermark and seed.

It does not make a video from a prompt alone (use flymyai/wan-3_0-t2v), does not continue a clip (use flymyai/wan-3_0-video-extend) and takes no first or last frame. Alibaba refuses videos and instructions its content inspection flags.

Billed per second of video at the resolution's rate, for the clip you give plus the clip you get: $0.05 at 480P, $0.10 at 720P, $0.20 at 1080P. A 5-second clip edited at 720P is 10 billed seconds, about $1.00; at 480P about $0.50. A refused or failed run costs nothing.

Price: From $0.05 per second of video, depending on resolution; the clip you give counts as well as the clip you get (30 s in all).

Run Alibaba Wan - video edit (3.0) via API

curl -X POST https://api.flymy.ai/api/v1/flymyai/wan-3_0-video-edit/predict \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Relight the scene as a warm golden sunset with long shadows. Keep everything else the same.","video":"https://storage.googleapis.com/open-inference-results/agent-files/1585/40ecb0fd73ed4b5eb3b34b53b3b7e290/bytedance-seedance-2_0-fast-text-to-video_a64b.mp4","resolution":"480P"}'

API documentation · Create a free account · Pricing

Related models