Wan
Alibaba's AI-powered audiovisual generation platform, specializing in Chinese language and style.
Visit Website ↗What is Wan
Wan (Tongyi Wansiang) is an AI audiovisual creation platform developed by Alibaba's Tongyi Laboratory, integrating text-to-image, image-to-video, text-to-video, and video editing functions into a single web page, wan.video, without requiring local environment setup. Its underlying Wan series model has gained significant attention in the open-source community, with version 2.1 ranking first in the VBench video generation benchmark, and subsequent versions 2.6 and 2.7 adding support for multi-camera narrative, native audio, and character consistency.
For creators in Taiwan and the Chinese-speaking region, Wan's most significant advantage is its understanding of Chinese. It excels in comprehending Chinese instructions and restoring Chinese-style scenes and aesthetics, making it a more suitable choice than European and American models, which often produce content with a sense of dissonance. When creating Chinese-style short videos or characters with Chinese dialogue, Wan is usually a more straightforward option.
Key Features and Use Cases
Wan's core capabilities cover text-to-video, image-to-video, and video editing, making it a one-stop platform for pre-production to post-production. The new version's addition of native audio and multi-camera functionality enables the creation of more complete video segments with camera movements, music, and sound effects, eliminating the need for post-production audio editing.
Suitable use cases include generating materials for social media short videos, creating Chinese-style visuals for advertising and e-commerce, and producing videos with Chinese dialogue for self-media creators or designers. Wan offers a free trial, with paid upgrades for advanced usage and high-resolution output. However, its bias towards Chinese-style content can be a double-edged sword – it may not be the best choice for European and American-style content.
Key Features
- All-in-one platform for text-to-image, image-to-video, and video editing
- High understanding and restoration of Chinese instructions and aesthetics
- Support for native audio and multi-camera narrative in new version
- Web-based, no local deployment required
- Underlying model is open-source, with top ranking in VBench video generation benchmark
Pros
- One of the best choices for Chinese content and dialogue
- Low barrier to entry, no download or configuration required
- Integrated video editing, streamlining pre-production to post-production
Cons
- Bias towards Chinese style, may not be the best choice for European and American-style content
- Limited free trial, paid upgrades required for advanced usage
- Video generation stability still affected by instructions and subject matter
Use Cases
- Quick generation of social media short video materials
- Creating Chinese-style visuals and storyboards for advertising and e-commerce
- Producing videos with Chinese dialogue
- Designers' creative experiments with image-to-video
Editor's Note
For Chinese creators, Wan's greatest value lies in its understanding of Chinese, saving time and effort in content creation. However, its bias towards Chinese style can be a double-edged sword. We give it a rating of 4.3.
FAQ
Does Wan really support Chinese better?
Yes, it has a better understanding of Chinese instructions and restores Chinese-style scenes and aesthetics more accurately than European and American models, making it a more suitable choice for Chinese content creation.
Do I need to pay to use Wan?
Wan offers a free trial, and paid upgrades are only required for advanced usage and high-resolution output. You can try it out for free before deciding to upgrade.