Overview

CogVideoX uses a 3D VAE and an expert transformer to generate coherent short videos. It was one of the first strong open video models that ran on a single GPU.

Highlights

  • 2B and 5B sizes
  • Text-to-video and image-to-video
  • 6-second clips at 720×480
  • Diffusers integration

Typical uses

Video generation research and prototyping.

Things to consider

  • Short, low-resolution clips compared with newer models.

Specifications are taken from the model card on Hugging Face and the provider’s announcement. Benchmark results depend on evaluation settings; see how AI benchmarks work.