Wan2Video - Multi-Shot Stories with Consistent Characters

Can Wan2Video maintain the same character across multiple shots in a single video?

Yes. Wan2Video is built to keep characters looking and sounding consistent across scenes within a 15-second video. You can create multi-shot narratives where the same character appears throughout without needing to regenerate or manually correct their appearance between takes.

What's the maximum length video Wan2Video can generate in one go?

Wan2Video generates videos up to 15 seconds long at 1080P resolution per generation. If you need longer content, you would need to create multiple clips and combine them in an editor afterward.

Does Wan2Video sync audio to video automatically or do you have to process them separately?

Wan2Video generates audio and video together in a single pass, so they arrive synchronized without needing separate audio processing. The tool aims for natural vocal tone and coherent audio-visual continuity, meaning lips should move with the spoken words.

Can you feed an existing video into Wan2Video to extend or modify it?

Yes. Wan2Video supports reference-based video generation, which means you can input an existing video and have it generate new content while preserving character identity and voice from your source material.