Alibaba unveils Wan2.1, the latest artificial intelligence model that can generate videos, consisting of 4 sub-models developed based on the Tongyi Wanxiang image creation model.
The 4 models differ in the number of parameters they can handle, including Wan2.1-T2V-14B, Wan2.1-I2V-14B-720P, Wan2.1-I2V-14B-480P, and the smallest model Wan2.1-T2V-1.3B, compatible with consumer-grade GPUs like RTX 4090.
Wan2.1 supports multiple input formats such as Text-to-Video, Image-to-Video, video editing, Text-to-Image, and even Video-to-Audio. Additionally, it can generate visual text output in both Chinese and English languages.
For more details and downloads, visit HuggingFace or GitHub.
Source: South China Morning Post and Alibaba
TLDR: Alibaba introduces Wan2.1, an AI model capable of creating various types of videos and visuals, available for download on HuggingFace and GitHub.
Leave a Comment