7-year-old graphics card GTX 1080 can run, picture generation video model Stable Video Diffusion update

CTOnews.com, Nov. 25, Stability AI released the Stable Video Diffusion model on Tuesday, which is based on the previously released Stable Diffusion text-to-picture model, which can be used to generate video images.

Stability AI recently updated the Stable Video Diffusion model to add ComfyUI tool support.

CTOnews.com Note: ComfyUI is the graphical user interface of Stable Diffusion, using a graphical / node interface to help users build complex workflows.

Stability AI said that after this update, users can use Nvidia GTX 1080 graphics cards with 8GB memory seven years ago (released in May 2016) to generate video with a resolution of 10240576 and a length of 25 frames.

The minimum threshold supported by AMD is 6800 XT. You can run ROCm in a Linux system. It takes about 3 minutes to generate video from an image.

In his blog post, Stability AI shares a sample process for calling Stable Video Diffusion in ComfyUI, one for a 14-frame model and one for a 25-frame model.

Stable Video Diffusion is currently in what Stability AI calls a "research preview" phase. People who want to use this model must agree to some terms of use that specify the expected application scenarios of Stable Video Diffusion (such as "educational or creative tools", "design and other artistic processes", etc.) and unexpected scenarios (such as "factual or true representation of people or events").