Which modes are available in Wan 2.6?

Wan 2.6 supports text-to-video, image-to-video, and video-to-video generation paths.

Can I use an existing video as input?

Yes. Video-to-video mode allows you to transform an input clip with new visual instructions.

It is a strong choice for hybrid pipelines where teams need to switch input types frequently.

Use concise prompts with explicit motion and scene constraints, and keep input media clear and stable.

A multi-mode generation stack for teams that need one model page to cover text, image, and video-driven workflows.

Mode

Prompt*

Duration

Resolution

Multi Shots

Credits are consumed when generation starts. Failed generations are refunded.

Preview Mode

Previewing: Wan 2.6 Sample

Mode

Prompt*

Duration

Resolution

Multi Shots

Credits are consumed when generation starts. Failed generations are refunded.

Preview Mode

Previewing: Wan 2.6 Sample

Use Wan 2.6 when your pipeline mixes multiple input types and you need one model endpoint for flexible generation tasks.

Three generation modes

Supports text-to-video, image-to-video, and video-to-video workflows.

Flexible production use

Useful for experimentation, adaptation, and remix-oriented video pipelines.

Single-page operations

Run mixed workflow generation from one consistent UI flow.

Modes

Text / Image / Video

Best for

Hybrid generation workflows

Quick answers about capabilities, input requirements, and output quality.