Wan 3.0: animate a still, then edit it properly

Wan 3.0 is an image-to-video model from Alibaba, available in REQO's AI Studio. It animates a still you upload rather than inventing a shot from a prompt alone. Clips run 4s to 10s, with a soundtrack generated alongside the picture, and it sits in the balanced tier at 4 credits for its shortest clip.

Against the 14 other image-to-video models in the studio it is the 4th-cheapest to run and comes back 3rd quickest, at 4 credits and about 60 seconds a generation.

In the picker it is described as: Newest Wan, native audio.

You already have the frame - a product shot, a still from a shoot, an image you generated a moment ago - and want up to 10s of movement out of it, at 720p, 480p or 1080p.

The clip you need fits inside 4s to 10s, and you can wait about 60s for it.

You need 16:9, 9:16 or 1:1 - no other image-to-video model here accepts as many natively.

You want sound generated with the picture instead of added on the timeline.

Where it runs out: Alibaba built this endpoint to animate a still rather than to generate one, so there is no prompt-only path into it. Clips stop at 10s; Seedance 1.5 Pro goes to 12s if you need one shot to run longer. Resolutions are limited to 720p, 480p or 1080p, and the higher ones cost more credits. At 4 credits it is not the cheap option: PixVerse V6 runs at 3.

Most generators hand you a clip and stop, which leaves you downloading files and re-uploading them somewhere else to make anything of them.

In REQO the Wan 3.0 result lands next to your screen recordings and uploads, about 60 seconds after you attach the still. Pick the aspect ratio before you generate (16:9, 9:16, 1:1); afterwards it is a crop.

The AI Studio (talking avatars, text-to-video and AI images) and the AI clip tools (background removal, chroma key, face touch-up, noise cancellation, auto cleanup, auto b-roll, auto-zoom) are Pro features. Generation includes 100 credits a month plus top-ups.