Wan 2.1 marks a significant breakthrough in the realm of video foundation models, setting fresh standards within the video production sector. By utilizing an advanced 3D VAE architecture coupled with sophisticated diffusion transformer technology, it achieves remarkable performance even on consumer-grade GPUs. Notably, this adaptable model excels in both text-to-video and image-to-video tasks and is the first to facilitate text generation in both English and Chinese.