Introducing Wan 3.0: Alibaba's Next-Generation Video Model
Wan 3.0 is a cutting-edge video generation model from Alibaba, now available in public beta. This advanced model is capable of producing high-quality video clips up to 30 seconds in length, all within a single pass. It supports a wide range of input modalities, including text, images, audio, video, and even documents, offering unparalleled flexibility for content creation.
Key Features:
- Extended Single-Pass Generation: Create complete scenes up to 30 seconds long without the need for stitching multiple clips, ensuring greater continuity and a smoother workflow.
- Omni-Modal Inputs: Utilize text prompts, reference images, video clips, audio files, and documents (doc, xls, ppt, pdf, md) to drive video generation.
- Fine-Grained Reference Control: Maintain consistency in characters, voices, props, spatial relationships, and overall style across generated shots.
- Multiple Resolutions: Output videos in 480P, 720P, or 1080P resolutions, with pricing scaling accordingly.
- Flexible Aspect Ratios: Supports various aspect ratios, including adaptive, 16:9, 9:16, 1:1, 4:3, and 3:4.
- Native Audio Track: Generated videos include a native audio track by default.
- Pay-as-you-go Pricing: Billed per second of generated video, scaled by resolution. Failed generations are automatically refunded.
Wan 3.0 is accessible via the reAPI platform, providing a unified endpoint for integrating this powerful video generation technology into your applications. Whether you're creating marketing content, social media clips, or narrative scenes, Wan 3.0 offers a robust and flexible solution.