Wan 2.5

Create cinematic AI videos from text or images

Generate video with Wan 2.5

Start from a still → add motion while keeping the subject

  1. Upload a keyframe or reference image in the Wan 2.5 playground above.
  2. Describe camera move, motion, and what must stay recognizable (face, product, logo).
  3. Generate a short clip, then refine the prompt if motion looks off.

Example prompts

  • Keep this character identical; gentle push-in camera; soft daylight.
  • Animate the product slowly rotating on the table; keep the label readable.

FAQ

Can I turn a photo into a video with Wan 2.5?

Yes — upload a still as the starting frame, then describe the motion you want.

How long are Wan 2.5 clips?

Choose 5s or 10s in the playground.

How do I keep the same person or product looking consistent?

Use a sharp, well-lit still and say what to preserve in the prompt (face, clothing, product silhouette). Stronger identity cues improve continuity, but results still vary by model.

How many credits does Wan 2.5 use?

Wan 2.5: 720p is 60 credits (5s) or 120 (10s); 1080p is 100 (5s) or 200 (10s). Check the Run button before you generate.

What aspect ratios does Wan 2.5 support?

Aspect ratio is not selectable here; pick 720p or 1080p resolution when offered.

Wan 2.5 – Image to Video | Image to Image AI