Wan 3.0 AI Video Generator

Wan 3.0 is the newest model in Alibaba's Wan series. It generates 3 to 30 second clips up to 1080P from a prompt alone, from a first frame, from first and last frames, or from a mix of references.

Reference-to-video accepts up to 10 assets and mixes types — images for identity and styling, video for motion and camera language, audio for voice and sound, plus document and webpage references for the content itself.

Wan 3.0 goes longer, up to 30 seconds, and takes the widest range of reference material. Wan 2.7 is the one to pick when you need to work on an existing clip, since it adds video extend and instruction-based video editing.

Anchor the model instead of relying on one vague prompt: supply reference images or start and end frames, say which reference controls identity and which controls style, and list what must stay unchanged — faces, props, brand text.

You can start for free. New accounts get bonus credits on signup, there is no credit card required, and the free plan has no daily generation limit. You can also claim extra credits every day by checking in. Everything runs in the browser on MakeFun AI — no GPU, no install, and no API key.

Yes. Videos generated on a paid MakeFun AI plan can be used commercially — ads, social media monetization, client deliverables and product marketing. You keep ownership of the output, with no watermark, no attribution requirement and no per-clip royalties.

Every finished job appears in the My Results tab beside the generator, where you can preview, download, share or delete it. The list keeps the last 3 days of generations, so download anything you want to keep before it rolls off.