On August 6, Zhipu AI announced that it will open source CogVideoX, a video generation model with the same origin as "Ying". According to reports, the CogVideoX open source model contains a number of different sizes of the model, the current open source CogvideOX-2B, it requires 18GB of video memory in FP-16 precision reasoning, fine tuning requires 40GB video memory, which means that a single 4090 video card can be inferred, A single A6000 video card can be fine-tuned.
Zhipu AI open source CogVideoX video generation model
On August 6, Zhipu AI announced that it will open source CogVideoX, a video generation model with the same origin as "Ying". According to reports, the CogVideoX open source model contains a number of different sizes of the model, the current open source CogvideOX-2B, it requires 18GB of video memory in FP-16 precision reasoning, fine tuning requires 40GB video memory, which means that a single 4090 video card can be inferred, A single A6000 video card can be fine-tuned.