zai-org/CogVideo

text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

52
/ 100
Established

Implements diffusion-based video generation with both SAT and Hugging Face Diffusers backends, supporting text-to-video, image-to-video, and video continuation tasks. Features a 3D Causal VAE for near-lossless video reconstruction and includes LoRA fine-tuning optimized for consumer GPUs (RTX 3060+). Integrates prompt optimization via LLM-4 and supports quantization through TorchAO for memory-efficient inference on consumer hardware.

12,515 stars.

No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

12,515

Forks

1,263

Language

Python

License

Apache-2.0

Last pushed

Nov 04, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/zai-org/CogVideo"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.