Skyreels V4

SkyReels-V4, developed by Skywork AI (Kunlun Tech), is the world’s first unified multimodal video foundation model that integrates video-audio co-generation, inpainting, and editing into one revolutionary architecture. Ranked #2 globally on the Artificial Analysis Text-to-Video (with Audio) Leaderboard, it outperforms industry giants and redefines what AI video creation can achieve. Official site: https://skyreels-v4.ai. If you want to know about more info, you can click there https://skyreels-v4.ai/blog.
Powered by a cutting-edge dual-stream MMDiT architecture with a shared MLLM text encoder, SkyReels-V4 understands and processes text, images, video clips, masks, and audio references simultaneously.