{"id":906,"date":"2025-07-06T05:28:51","date_gmt":"2025-07-06T05:28:51","guid":{"rendered":"https:\/\/janusai.pro\/?p=906"},"modified":"2025-07-06T05:28:52","modified_gmt":"2025-07-06T05:28:52","slug":"janus-4o-sharegpt-4o-image","status":"publish","type":"post","link":"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/","title":{"rendered":"Den nya stj\u00e4rnan inom multimodal bildgenerering: Janus-4o? DelaGPT-4o-Image s\u00e4tter en ny standard f\u00f6r datam\u00e4ngder och anpassar bildgenerering till GPT-4o."},"content":{"rendered":"<div style=\"margin-top: 0px; margin-bottom: 0px;\" class=\"sharethis-inline-share-buttons\" ><\/div>\n<p><a href=\"https:\/\/sharegpt4o.github.io\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">DelaGPT-4o-bild<\/a> \u00e4r en storskalig, h\u00f6gkvalitativ bildgenereringsdataupps\u00e4ttning d\u00e4r alla bilder genereras med hj\u00e4lp av GPT-4os bildgenereringsfunktioner.<\/p>\n\n\n\n<p>Denna dataset syftar till att kombinera f\u00f6rdelarna med multimodala modeller med \u00f6ppen k\u00e4llkod med GPT-4os styrkor inom skapande av visuellt inneh\u00e5ll. <\/p>\n\n\n\n<p>Den inneh\u00e5ller 45 000 text-till-bild- och 46 000 bild-till-text-exempel, vilket g\u00f6r den till en praktisk resurs f\u00f6r att f\u00f6rb\u00e4ttra multimodala modeller vid bildgenerering och redigering.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"998\" height=\"700\" src=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/f48c8349-9310-48a1-9276-d7614aa958d9.png\" alt=\"\" class=\"wp-image-911\" srcset=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/f48c8349-9310-48a1-9276-d7614aa958d9.png 998w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/f48c8349-9310-48a1-9276-d7614aa958d9-300x210.png 300w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/f48c8349-9310-48a1-9276-d7614aa958d9-768x539.png 768w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/f48c8349-9310-48a1-9276-d7614aa958d9-18x12.png 18w\" sizes=\"auto, (max-width: 998px) 100vw, 998px\" \/><\/figure>\n\n\n\n<p>Janus-4o \u00e4r en multimodal LLM som kan generera text-till-bild och text+bild-till-bild. Den \u00e4r baserad p\u00e5 Janus-Pro och finjusterad med hj\u00e4lp av datasetet ShareGPT-4o-Image. J\u00e4mf\u00f6rt med Janus-Pro introducerar Janus-4o text+bild-till-bild-genereringsfunktioner och uppn\u00e5r betydande f\u00f6rb\u00e4ttringar i text-till-bild-generering.<\/p>\n\n\n\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_82_2 counter-hierarchy ez-toc-counter ez-toc-grey ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Inneh\u00e5llsf\u00f6rteckning<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"V\u00e4xla inneh\u00e5llsf\u00f6rteckning\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewbox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewbox=\"0 0 24 24\" version=\"1.2\" baseprofile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/#Dataset_Overview\" >\u00d6versikt \u00f6ver datam\u00e4ngden<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/#Related_Links\" >Relaterade l\u00e4nkar<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/#Paper_Introduction\" >Introduktion till dokumentet<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/#Method_Overview\" >Metod\u00f6versikt<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/#Experimental_Results\" >Experimentella resultat<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/janusai.pro\/sv\/janus-4o-sharegpt-4o-image\/#Conclusions\" >Slutsatser<\/a><\/li><\/ul><\/nav><\/div>\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Dataset_Overview\"><\/span>\u00d6versikt \u00f6ver datam\u00e4ngden<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Datasetet ShareGPT-4o-Image inneh\u00e5ller 91 000 exempel p\u00e5 GPT-4o-bilder, kategoriserade enligt f\u00f6ljande:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Text-till-bild: 45 717<\/li>\n\n\n\n<li>Text-plus-bild-till-bild: 46 539<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Related_Links\"><\/span>Relaterade l\u00e4nkar<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Koda: <a href=\"https:\/\/github.com\/FreedomIntelligence\/ShareGPT-4o-Image\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">github klicka h\u00e4r<\/a><\/p>\n\n\n\n<p>Modell: <a href=\"https:\/\/huggingface.co\/FreedomIntelligence\/Janus-4o-7B\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">h\u00e4mta ShareGPT-4o-Image-modellen<\/a><\/p>\n\n\n\n<p>Papper: <a href=\"https:\/\/arxiv.org\/pdf\/2506.18095\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">klicka h\u00e4r<\/a><\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Paper_Introduction\"><\/span>Introduktion till dokumentet<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Nya framsteg inom multimodala genereringsmodeller har \u00f6ppnat upp f\u00f6r realistisk, instruktionsanpassad bildgenerering. Ledande system som GPT-4o-Image \u00e4r dock fortfarande propriet\u00e4ra och o\u00e5tkomliga.<\/p>\n\n\n\n<p>F\u00f6r att g\u00f6ra dessa funktioner tillg\u00e4ngliga f\u00f6r allm\u00e4nheten introducerar artikeln ShareGPT-4o-Image, den f\u00f6rsta datam\u00e4ngden som inneh\u00e5ller 45 000 text-till-bild- och 46 000 text-plus-bild-till-bild-exempel, alla syntetiserade med hj\u00e4lp av GPT-4os bildgenereringsfunktioner f\u00f6r att f\u00f6rfina dess avancerade bildgenereringsf\u00f6rm\u00e5ga. Med hj\u00e4lp av denna datam\u00e4ngd utvecklade artikeln Janus-4o, en multimodal stor spr\u00e5kmodell som kan generera text-till-bild och text-plus-bild-till-bild.<\/p>\n\n\n\n<p>Janus-4o f\u00f6rb\u00e4ttrar inte bara text-till-bild-genereringsfunktionerna avsev\u00e4rt j\u00e4mf\u00f6rt med f\u00f6reg\u00e5ngaren Janus-Pro, utan introducerar \u00e4ven text-plus-bild-till-bild-genereringsfunktioner. Det \u00e4r v\u00e4rt att notera att den uppn\u00e5r imponerande prestanda n\u00e4r det g\u00e4ller att generera bilder fr\u00e5n text och bilder fr\u00e5n grunden med endast 91 000 syntetiska samplingar och tr\u00e4nad i 6 timmar p\u00e5 en 8\u00d7A800 GPU-maskin.<\/p>\n\n\n\n<p>Vi hoppas att lanseringen av ShareGPT-4o-Image och Janus-4o kommer att fr\u00e4mja \u00f6ppen forskning inom fotorealistisk, instruktionsanpassad bildgenerering.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Method_Overview\"><\/span>Metod\u00f6versikt<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1028\" height=\"718\" src=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/74bd55e5-5cc6-49e8-be21-cf5c4a66042d.png\" alt=\"\" class=\"wp-image-908\" srcset=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/74bd55e5-5cc6-49e8-be21-cf5c4a66042d.png 1028w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/74bd55e5-5cc6-49e8-be21-cf5c4a66042d-300x210.png 300w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/74bd55e5-5cc6-49e8-be21-cf5c4a66042d-1024x715.png 1024w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/74bd55e5-5cc6-49e8-be21-cf5c4a66042d-768x536.png 768w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/74bd55e5-5cc6-49e8-be21-cf5c4a66042d-18x12.png 18w\" sizes=\"auto, (max-width: 1028px) 100vw, 1028px\" \/><\/figure>\n\n\n\n<p><strong>ShareGPT-4o-Image f\u00f6rb\u00e4ttrar prestandan f\u00f6r bildgenerering.<\/strong> Genom att finjustera Janus-Pro med ShareGPT-4o-Image genererade vi Janus-4o, vilket visar avsev\u00e4rt f\u00f6rb\u00e4ttrad bildgenereringsprestanda. Janus-4o st\u00f6der \u00e4ven text-till-bild- och bild-till-bild-generering, vilket \u00f6vertr\u00e4ffar andra riktm\u00e4rken med endast 91 000 tr\u00e4ningsprover.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"370\" src=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/fc3b163f-d1d2-42f5-81bc-884eb677ea52.png\" alt=\"\" class=\"wp-image-910\" srcset=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/fc3b163f-d1d2-42f5-81bc-884eb677ea52.png 1024w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/fc3b163f-d1d2-42f5-81bc-884eb677ea52-300x108.png 300w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/fc3b163f-d1d2-42f5-81bc-884eb677ea52-768x278.png 768w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/fc3b163f-d1d2-42f5-81bc-884eb677ea52-18x7.png 18w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><strong>\u00d6versikt \u00f6ver Janus-4o-modellen.<\/strong> Modellen \u00e4r baserad p\u00e5 Janus-Pro och konstruerad genom finjustering p\u00e5 ShareGPT-4o-Image. Den inneh\u00e5ller f\u00f6rb\u00e4ttringar f\u00f6r att st\u00f6dja text-till-bild- och bild-till-bild-generering. B\u00e5de text-till-bild- och text-till-bild-uppgifter tr\u00e4nas gemensamt.<\/p>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1058\" height=\"304\" src=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/2b81408d-3c8b-45a8-ac73-ee0a48164c05.png\" alt=\"\" class=\"wp-image-909\" srcset=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/2b81408d-3c8b-45a8-ac73-ee0a48164c05.png 1058w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/2b81408d-3c8b-45a8-ac73-ee0a48164c05-300x86.png 300w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/2b81408d-3c8b-45a8-ac73-ee0a48164c05-1024x294.png 1024w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/2b81408d-3c8b-45a8-ac73-ee0a48164c05-768x221.png 768w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/2b81408d-3c8b-45a8-ac73-ee0a48164c05-18x5.png 18w\" sizes=\"auto, (max-width: 1058px) 100vw, 1058px\" \/><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Experimental_Results\"><\/span>Experimentella resultat<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<figure class=\"wp-block-image\"><img loading=\"lazy\" decoding=\"async\" width=\"1072\" height=\"1140\" src=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/72720ada-7418-4979-a8fd-4ce09050d696.png\" alt=\"\" class=\"wp-image-907\" srcset=\"https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/72720ada-7418-4979-a8fd-4ce09050d696.png 1072w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/72720ada-7418-4979-a8fd-4ce09050d696-282x300.png 282w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/72720ada-7418-4979-a8fd-4ce09050d696-963x1024.png 963w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/72720ada-7418-4979-a8fd-4ce09050d696-768x817.png 768w, https:\/\/janusai.pro\/wp-content\/uploads\/2025\/07\/72720ada-7418-4979-a8fd-4ce09050d696-11x12.png 11w\" sizes=\"auto, (max-width: 1072px) 100vw, 1072px\" \/><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Conclusions\"><\/span>Slutsatser<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>ShareGPT-4o-Image \u00e4r den f\u00f6rsta storskaliga datam\u00e4ngden som kan f\u00e5nga GPT-4os avancerade bildgenereringsfunktioner inom text-till-bild- och text-till-bild-generering. Baserat p\u00e5 denna datam\u00e4ngd utvecklade artikeln Janus-4o, en maskininl\u00e4rningsmodell (MLLM) som kan generera h\u00f6gkvalitativa bilder fr\u00e5n ren text eller bild-text-kombinationer.<\/p>\n\n\n\n<p>Janus-4o uppn\u00e5r betydande f\u00f6rb\u00e4ttringar i text-till-bild-generering och mycket konkurrenskraftiga resultat i text-till-bild-uppgifter, vilket demonstrerar den h\u00f6ga kvaliteten och praktiska anv\u00e4ndbarheten hos ShareGPT-4o-Image.<\/p>\n\n\n\n<p>Tack vare effektiviteten hos sj\u00e4lvregressiv bildgenerering baserad p\u00e5 MLLM kan Janus-4o tr\u00e4nas p\u00e5 bara 6 timmar p\u00e5 en 8\u00d7A800 GPU-maskin och uppn\u00e5r betydande prestandaf\u00f6rb\u00e4ttringar med extremt l\u00e5ga ber\u00e4kningskrav.<\/p>","protected":false},"excerpt":{"rendered":"<p>ShareGPT-4o-Image \u00e4r en storskalig, h\u00f6gkvalitativ bildgenereringsdataupps\u00e4ttning d\u00e4r alla bilder genereras med hj\u00e4lp av GPT-4os bildgenereringsfunktioner. Denna dataupps\u00e4ttning syftar till att kombinera f\u00f6rdelarna med multimodala modeller med \u00f6ppen k\u00e4llkod med GPT-4os styrkor inom skapande av visuellt inneh\u00e5ll. Den inneh\u00e5ller 45 000 text-till-bild- och 46 000 bild-till-text-exempel, vilket g\u00f6r den till en praktisk resurs f\u00f6r att f\u00f6rb\u00e4ttra multimodala modeller inom bild\u2026<\/p>","protected":false},"author":2,"featured_media":859,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_kadence_starter_templates_imported_post":false,"_kad_post_transparent":"","_kad_post_title":"","_kad_post_layout":"","_kad_post_sidebar_id":"","_kad_post_content_style":"","_kad_post_vertical_padding":"","_kad_post_feature":"","_kad_post_feature_position":"","_kad_post_header":false,"_kad_post_footer":false,"footnotes":""},"categories":[1],"tags":[],"class_list":["post-906","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/posts\/906","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/comments?post=906"}],"version-history":[{"count":1,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/posts\/906\/revisions"}],"predecessor-version":[{"id":912,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/posts\/906\/revisions\/912"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/media\/859"}],"wp:attachment":[{"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/media?parent=906"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/categories?post=906"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/janusai.pro\/sv\/wp-json\/wp\/v2\/tags?post=906"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}