{"id":3955,"date":"2024-02-16T08:47:37","date_gmt":"2024-02-16T00:47:37","guid":{"rendered":"https:\/\/www.1ai.net\/?p=3955"},"modified":"2024-02-16T08:47:37","modified_gmt":"2024-02-16T00:47:37","slug":"stability-ai-%e6%8e%a8%e5%87%ba%e6%96%b0%e4%b8%80%e4%bb%a3%e6%96%87%e7%94%9f%e5%9b%be%e6%a8%a1%e5%9e%8b-stable-cascade%ef%bc%8c%e5%8f%b7%e7%a7%b0%e6%af%94-sdxl-%e6%9b%b4%e9%ab%98%e6%95%88%e6%9b%b4","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/3955.html","title":{"rendered":"Stability AI launches a new generation of text graph model Stable Cascade, claiming to be more efficient and powerful than SDXL"},"content":{"rendered":"

according to Stability AI<\/a> Official press release, Stability AI recently launched a new generation ofWenshengtu Model<\/a>, the model is built on the W\u00fcrstchen architecture and claims to be easy to train and fine-tune on consumer-grade hardware.<\/p>\n

\"1b34ebba-0c5a-4698-9461-56a4cb966ec3\"<\/p>\n

\u25b2 Image source: Stability AI official press release (the same below)<\/p>\n

Officials claim that compared to the SDXL familiar to the industry,The new Stable Cascade model has improved performance and claimed content quality<\/span>Currently, the relevant data of the Stable Cascade model has been made public on the GitHub page, but only non-commercial use is allowed.<\/p>\n

IT House noted that after the user input section, the content would be converted from the Stable Cascade model to a small data set of 24 x 24 sizes, the model would decode the small image data to generate pictures and continue to zoom in the picture to high resolution images, so that additional training and fine-tuning of various aspects of the model could be undertaken as a result of the separation of the series of steps\u3002<\/p>\n

\"61073a73-5f08-4541-8e9a-470315ae9539\"<\/p>\n

Stability AI said that because the Stable Cascade model adopts such a "modular" design,Therefore, it can effectively reduce the video memory used for inference, claiming that only 20 GB of video memory is needed to run<\/span>.<\/p>\n

Stability AI also compared the Stable Cascade model with other industry competitors such as Playground v2, SDXL, SDXL Turbo, and W\u00fcrstchen v2, claiming that Stable Cascade is "almost always the best performing model" in terms of prompt alignment and generated image details. In terms of inference speed, even though the largest Stable Cascade model has 1.4 billion more parameters than Stable Diffusion XL, it still has a faster inference speed.<\/p>\n

\"4198b4da-c6be-4105-a2aa-d1441ac8f65a\"<\/p>\n

Based on this, the official believes that Stable Cascade has better architectural design and can maintain efficient inference speed while maintaining high-quality output.<\/p>\n

Model-generated content pictures:<\/h3>\n

\"b189a807-1dd6-4e4e-b659-66e43257e2da\"<\/p>","protected":false},"excerpt":{"rendered":"

According to an official Stability AI press release, Stability AI has launched a new generation of graph models called Stable Cascade, which is built on the W\u00fcrstchen architecture and claims to be easy to train and fine-tune on consumer-grade hardware. \u25b2 Source Stability AI official press release (below) Officials claim that the new Stable Cascade model has improved performance and quality of claimed content compared to the industry-familiar SDXL, and that data related to the Stable Cascade model has been made available on the GitHub page for non-commercial use only. IT House notes that<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[718,1245],"collection":[],"class_list":["post-3955","post","type-post","status-publish","format-standard","hentry","category-news","tag-stability-ai","tag-1245"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/3955","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=3955"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/3955\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=3955"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=3955"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=3955"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=3955"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}