LightX2V

Industry's first open-source inference framework enabling real-time video generation. Supports text-to-video, image-to-video, text-to-image, and image editing tasks. Through DiT distillation acceleration, lightweight VAE, and sparse attention, generates a 5-second video on consumer GPUs in under 5 seconds. Over 5 million downloads. Powers SenseTime's Seko 2.0 drama generation platform.