Shulou(Shulou.com)11/24 Report--

CTOnews.com August 2 news, Nvidia researchers recently launched a "Wensheng map" model called Perfusion, the biggest feature of the model is "small," the model volume is only 100KB, training time is only 4 minutes, but the generated pictures are still remarkable.

It is reported that Nvidia researchers in cooperation with Tel Aviv University in Israel launched the model, officials said, Perfusion model generated image quality and Stability AI Stable Diffusion v1.5 version is roughly the same, and in terms of efficiency, thanks to its lightweight volume, Perfusion model efficiency in the industry in the "first echelon."

In addition, the Perfusion model's ability to "undertake the above" is also quite outstanding, users can gradually guide the model distribution for image creation through text prompts, Nvidia said, Perfusion's accurate value can match other competitors in the industry, bringing "excellent visual effect" to the generated images.

CTOnews.com found that Nvidia's research team currently only publishes papers on its website, but the research team promises to "publish code soon." Nvidia's stock has risen 230% this year, and this lightweight Perfusion model is expected to give Nvidia even more advantages, thus continuing to increase market attention for Nvidia.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.