Wan 2.1 represents a groundbreaking collection of video foundation models that redefine the benchmarks for video creation. By leveraging a sophisticated 3D VAE architecture together with enhanced diffusion transformer technology, it achieves exceptional performance on standard consumer-grade GPUs. This versatile model enables both text-to-video and image-to-video capabilities, making it the first to feature text generation in both English and Chinese.
.png?width=64)
