Wan 2.2
By Alibaba Tongyi Lab
Wan 2.2 is now available on WaveSpeedAI!
Open and Advanced Large-Scale Video Generative Models
🚀 Effective MoE Architecture
Advanced Mixture-of-Experts architecture separates denoising process with specialized expert models, enlarging model capacity while maintaining computational cost
🎯 Cinematic-level Aesthetics
Meticulously curated aesthetic data with detailed labels for lighting, composition, contrast, and color tone for precise and controllable cinematic style generation
🌟 Complex Motion Generation
Trained on +65.6% more images and +83.2% more videos, achieving superior performance in motions, semantics, and aesthetics
Model Downloads
Text to Video Models
Image to Video Models
Efficient High-Definition Hybrid TI2V
5B model with advanced Wan2.2-VAE achieving 16×16×4 compression ratio. Supports both text-to-video and image-to-video generation at 720P resolution with 24fps.