Wan 2.1 VACE Inpainting

Wan 2.1 VACE Inpainting is a Wan-family route for masked video inpainting. Wan’s video-focused generative architecture models appearance and time together, so supplied references, structural conditions, masks, speech, or written scene direction can influence a coherent sequence rather than a collection of unrelated frames. This exact route concentrates that architecture on its named production task.

Use the route when the available source material and final brief specifically require masked video inpainting. Identify every subject or condition, explain the intended action or transformation, and describe camera behavior, pacing, environment, lighting, atmosphere, and preservation constraints. This separates creative intent from the local duration, resolution, and delivery fields documented below.

Highlights

• Temporal hole filling. Reconstructs masked video regions coherently across time.
• Motion-aware boundary blending. Matches movement, texture, perspective, and illumination around the edited area.
• Prompt-directed replacement. Introduces requested content inside the mask while respecting the source clip.
• Unmasked-region preservation. Keeps areas outside the edit mask connected to the original sequence.

Pricing

When to Use

Prompt Guide

Identify the primary subject and every source or condition, state the intended transformation or action, then describe composition, camera or viewpoint, lighting, materials, pacing, atmosphere, and exact preservation requirements. Refer to multiple inputs in their schema order.

{
  "prompt": "A precisely directed composition with explicit subject, transformation, camera or viewpoint, lighting, material, and preservation requirements",
  "mask": "https://example.com/reference.png",
  "seed": 1,
  "video": "https://example.com/source.mp4"
}

Technical Specs

Related Models

• alibaba/wan/2.1/text-to-video
• alibaba/wan/2.1/image-to-video
• alibaba/wan/2.1/vace