WanAI Releases Wan2.1: Efficient Video Generation Model

WanAI introduces Wan2.1, an efficient video generation model supporting text-to-video, image-to-video, and other tasks, with powerful video encoding and decoding capabilities.

WanAI has released its latest video generation model, Wan2.1. The model is available on Hugging Face and offers a variety of features, including text-to-video (T2V), image-to-video (I2V), video editing, text-to-image (T2I), and video-to-audio (V2A). Wan2.1 is licensed under Apache 2.0, requires 8.19GB of VRAM, and runs on most GPUs. It supports both Chinese and English text generation and features a powerful video VAE that can encode/decode 1080P videos with temporal precision. Additionally, research suggests that 'wan2.1 alibabacloud' likely refers to Alibaba Cloud's Cloud WAN service, a network solution for global connectivity.