{"id":4564,"date":"2024-02-28T09:29:05","date_gmt":"2024-02-28T01:29:05","guid":{"rendered":"https:\/\/www.1ai.net\/?p=4564"},"modified":"2024-02-28T09:29:05","modified_gmt":"2024-02-28T01:29:05","slug":"%e8%b0%b7%e6%ad%8c-deepmind-%e6%8e%a8%e5%87%ba-genie-%e6%a8%a1%e5%9e%8b%ef%bc%9a110-%e4%ba%bf%e4%b8%aa%e5%8f%82%e6%95%b0%ef%bc%8c%e5%9f%ba%e4%ba%8e%e5%9b%be%e7%89%87%e5%92%8c%e6%8f%90%e7%a4%ba","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/4564.html","title":{"rendered":"Google DeepMind launches Genie model: 11 billion parameters, generates 2D games based on pictures and prompts"},"content":{"rendered":"<p data-vmark=\"519c\"><a href=\"https:\/\/www.1ai.net\/en\/tag\/%e8%b0%b7%e6%ad%8c\" title=\"[View articles tagged with [Google]]\" target=\"_blank\" >Google<\/a>Its <a href=\"https:\/\/www.1ai.net\/en\/tag\/deepmind\" title=\"_Other Organiser\" target=\"_blank\" >DeepMind<\/a> The team recently launched <a href=\"https:\/\/www.1ai.net\/en\/tag\/genie\" title=\"[See articles with [Genie] labels]\" target=\"_blank\" >Genie<\/a> brand new <a href=\"https:\/\/www.1ai.net\/en\/tag\/ai%e6%a8%a1%e5%9e%8b\" title=\"[View articles tagged with [AI models]]\" target=\"_blank\" >AI Models<\/a>,<strong>With 11 billion parameters, it is capable of generating a more complete 2D platformer based on a single image and prompts entered by the user.<\/strong><\/p>\n<p data-vmark=\"33e7\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-4565\" title=\"64d10543-25e8-47c9-a452-2fbbfbd983fc\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/02\/64d10543-25e8-47c9-a452-2fbbfbd983fc.jpg\" alt=\"64d10543-25e8-47c9-a452-2fbbfbd983fc\" width=\"720\" height=\"823\" \/><\/p>\n<p data-vmark=\"1a9f\">Genie has been trained on over 200,000 hours of 2D gameplay videos from around the web, and is currently only a research preview of games that are more like 2D platformers than full-blown VR games.<\/p>\n<p data-vmark=\"6852\">The team commented, \"Genie has 11B parameters and can be described as a base world model. It consists of a spatio-temporal video tagger, an autoregressive dynamics model, and a simple and extensible potential action model.\"<\/p>\n<p data-vmark=\"182b\">Genie is trained on unsupervised and unlabeled video, and is therefore able to learn the movements, controls, and actions of a wide variety of characters in a consistent manner.<\/p>\n<p data-vmark=\"c0b8\">For users, basically just provide a sketch on paper, a perfect piece of digital art, or even an AI-generated description of the 2D world, and Genie will do the rest, generating the 2D game for you.<\/p>\n<p data-vmark=\"0a89\">It generates images and other assets, turns your sketch into a fully realized open world, and then predicts the next pixel frame based on player-provided actions.<\/p>","protected":false},"excerpt":{"rendered":"<p>Google's DeepMind team recently unveiled Genie, a new AI model with 11 billion parameters capable of generating a more complete 2D platformer based on a single image and prompts entered by the user. Genie was trained on more than 200,000 hours of 2D gameplay videos from around the web, and is currently only a research preview of games that are more like 2D platformers than full-blown VR games. The team commented, \"Genie has 11B parameters and can be described as a base world model. It consists of a spatio-temporal video tagger, an autoregressive dynamics model, and a simple and extensible latent action model.\" Genie is in the context of unsupervised and unlabeled video<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[167,593,1390,281],"collection":[],"class_list":["post-4564","post","type-post","status-publish","format-standard","hentry","category-news","tag-ai","tag-deepmind","tag-genie","tag-281"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/4564","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=4564"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/4564\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=4564"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=4564"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=4564"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=4564"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}