Wan-2.1 First-Last-Frame-to-Video

Wan-2.1 First-Last-Frame-to-Video is a Wan-family route for first-and-last-frame video synthesis. 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 first-and-last-frame video synthesis. 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

• Endpoint-constrained storytelling. Generates the motion that connects supplied opening and closing frames.
• Transition continuity. Builds a plausible temporal path between two visual endpoints.
• Subject and scene consistency. Keeps recognizable content coherent while pose, camera, or environment evolves.
• Prompt-guided motion. Uses written action and camera direction to shape how the transition unfolds.

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",
  "image": "https://example.com/reference.png",
  "seed": 1,
  "end_image": "https://example.com/reference.png",
  "resolution": "480p"
}

Technical Specs

Related Models

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