{"id":234598,"date":"2024-04-09T06:30:00","date_gmt":"2024-04-09T06:30:00","guid":{"rendered":"https:\/\/datascientest.com\/?p=234598"},"modified":"2024-04-03T09:25:05","modified_gmt":"2024-04-03T09:25:05","slug":"stable-diffusion-tout-savoir","status":"publish","type":"post","link":"https:\/\/datascientest.com\/stable-diffusion-tout-savoir","title":{"rendered":"Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"234598\" class=\"elementor elementor-234598\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-f76e126 e-flex e-con-boxed e-con\" data-id=\"f76e126\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-addb093 elementor-widget elementor-widget-heading\" data-id=\"addb093\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-heading-title{padding:0;margin:0;line-height:1}.elementor-widget-heading .elementor-heading-title[class*=elementor-size-]>a{color:inherit;font-size:inherit;line-height:inherit}.elementor-widget-heading .elementor-heading-title.elementor-size-small{font-size:15px}.elementor-widget-heading .elementor-heading-title.elementor-size-medium{font-size:19px}.elementor-widget-heading .elementor-heading-title.elementor-size-large{font-size:29px}.elementor-widget-heading .elementor-heading-title.elementor-size-xl{font-size:39px}.elementor-widget-heading .elementor-heading-title.elementor-size-xxl{font-size:59px}<\/style><h2 class=\"elementor-heading-title elementor-size-default\">La startup londonienne stability.ai a r\u00e9ussi l\u2019exploit d\u2019imposer une IA g\u00e9n\u00e9rative d\u2019images face \u00e0 deux mastodontes am\u00e9ricains. L\u2019application s\u2019appelle Stable Diffusion et elle a bien des atouts\u2026<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4cb1fd6 elementor-widget elementor-widget-text-editor\" data-id=\"4cb1fd6\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-text-editor.elementor-drop-cap-view-stacked .elementor-drop-cap{background-color:#69727d;color:#fff}.elementor-widget-text-editor.elementor-drop-cap-view-framed .elementor-drop-cap{color:#69727d;border:3px solid;background-color:transparent}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap{margin-top:8px}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap-letter{width:1em;height:1em}.elementor-widget-text-editor .elementor-drop-cap{float:left;text-align:center;line-height:1;font-size:50px}.elementor-widget-text-editor .elementor-drop-cap-letter{display:inline-block}<\/style>\t\t\t\t<p><span style=\"font-weight: 400;\">Si l\u2019on questionne le grand public sur l\u2019intelligence artificielle (IA), deux noms surgissent spontan\u00e9ment. Le premier est bien \u00e9videmment ChatGPT, lequel domine tr\u00e8s largement le march\u00e9 de l\u2019IA. Toutefois, il n\u2019est pas rare d\u2019entendre \u00e9galement la mention de Midjourney. Bien que la prise en main de cette IA g\u00e9n\u00e9rative d\u2019images ne soit pas ais\u00e9e, Midjourney a attir\u00e9 des millions d\u2019utilisateurs.<\/span><\/p><p><span style=\"font-weight: 400;\">Midjourney dispose toutefois de deux concurrents principaux :<\/span><\/p><ul><li><span style=\"font-weight: 400;\"><a href=\"https:\/\/datascientest.com\/dall-e-etiquette-ses-creations-le-debut-dune-ere-de-confiance\">Dall.e<\/a> 3 de <a href=\"https:\/\/datascientest.com\/open-ai-atteindra-un-chiffre-daffaires-record-dans-les-12-prochains-mois\">OpenAI<\/a>, cr\u00e9ateur de <a href=\"https:\/\/datascientest.com\/chatgpt-4-tout-savoir\">ChatGPT<\/a> qui a l&rsquo;avantage d&rsquo;\u00eatre accessible depuis la version payante de ChatGPT mais aussi depuis le moteur de recherche Bing de Microsoft.<\/span><\/li><li><span style=\"font-weight: 400;\">Stable Diffusion, une \u0153uvre du\u00a0 londonien, Stability.ai\u2026<\/span><\/li><\/ul>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-6a2eea8 e-flex e-con-boxed e-con\" data-id=\"6a2eea8\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-a011e3c elementor-widget elementor-widget-heading\" data-id=\"a011e3c\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Comment \u00e7a marche une IA g\u00e9n\u00e9rative d\u2019images ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5069e87 elementor-widget elementor-widget-text-editor\" data-id=\"5069e87\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Pour g\u00e9n\u00e9rer une image, il faut partir d\u2019un <a href=\"https:\/\/datascientest.com\/prompt-tout-savoir\">prompt<\/a>, comme dans ChatGPT. Si l\u2019on manque d\u2019inspiration, on peut s\u2019inspirer de cr\u00e9ations d\u00e9j\u00e0 r\u00e9alis\u00e9es en copiant leur prompt. Ainsi, Stable Diffusion produit des <strong>images d\u2019une qualit\u00e9 digne d&rsquo;un graphiste ou un photographe professionnel<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Le r\u00e9sultat correspond rarement \u00e0 ce que l\u2019on attendait et donc on prend l\u2019habitude de sp\u00e9cifier aussi des \u00ab <strong>negatives prompts<\/strong> \u00bb, en clair, d\u2019indiquer aussi ce que l\u2019on ne souhaite pas voir appara\u00eetre dans l\u2019image : pas de chien, pas de moustache, etc.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-950349a elementor-widget elementor-widget-heading\" data-id=\"950349a\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h4 class=\"elementor-heading-title elementor-size-medium\">Autres capacit\u00e9s<\/h4>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-802193e elementor-widget elementor-widget-text-editor\" data-id=\"802193e\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Stable Diffusion n\u2019est pas juste bon \u00e0 cr\u00e9er des images. Il peut \u00e9galement partir d\u2019une image existante et proposer des variations de celle-ci. Nous pouvons aussi lui confier d&rsquo;autres t\u00e2ches telles que l&rsquo;<strong>inpainting<\/strong> (reconstruction d\u2019images d\u00e9t\u00e9rior\u00e9es), l&rsquo;<strong>outpainting<\/strong> (\u00e9tendre une image en inventant les parties manquantes), etc.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3d1ef3e elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"3d1ef3e\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formation-prompt-engineering\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Apprendre \u00e0 faire des Prompts efficaces<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-c33feac e-flex e-con-boxed e-con\" data-id=\"c33feac\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-0e1c6d2 elementor-widget elementor-widget-heading\" data-id=\"0e1c6d2\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Comment fonctionne une IA g\u00e9n\u00e9rative ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-cc93c44 elementor-widget elementor-widget-text-editor\" data-id=\"cc93c44\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">L\u2019<a href=\"https:\/\/datascientest.com\/2024-sera-un-tournant-pour-lia-generative\">IA g\u00e9n\u00e9rative<\/a> repose sur deux technologies essentielles : <a href=\"https:\/\/datascientest.com\/quelle-difference-entre-le-machine-learning-et-deep-learning\">le machine learning, et le deep learning<\/a>, lequel repose sur des mod\u00e8les math\u00e9matiques plus \u00e9labor\u00e9s. Dans les deux cas de figure, l\u2019application analyse de dizaines de millions d\u2019images au sein desquelles,<strong> l\u2019IA est entra\u00een\u00e9e \u00e0 identifier des \u00e9l\u00e9ments<\/strong> : personnes, chiens et chats, b\u00e2timents, plantes, appareils\u2026<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5a3a2e0 elementor-widget elementor-widget-image\" data-id=\"5a3a2e0\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-image{text-align:center}.elementor-widget-image a{display:inline-block}.elementor-widget-image a img[src$=\".svg\"]{width:48px}.elementor-widget-image img{vertical-align:middle;display:inline-block}<\/style>\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"875\" height=\"500\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion_generative_ai.jpg\" class=\"attachment-full size-full wp-image-234628\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion_generative_ai.jpg 875w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion_generative_ai-300x171.jpg 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion_generative_ai-768x439.jpg 768w\" sizes=\"(max-width: 875px) 100vw, 875px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-c96d526 e-flex e-con-boxed e-con\" data-id=\"c96d526\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-a1c8c2b elementor-widget elementor-widget-heading\" data-id=\"a1c8c2b\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Un peu d\u2019histoire<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5535412 elementor-widget elementor-widget-text-editor\" data-id=\"5535412\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\"><strong>Emad Mostaque<\/strong>, fondateur de Stability AI a fond\u00e9 sa soci\u00e9t\u00e9 \u00e0 Londres en fin d&rsquo;ann\u00e9e 2020, et donc deux ans avant l\u2019explosion de ChatGPT. Autant dire qu\u2019il a \u00e9t\u00e9 visionnaire.<\/span><\/p><p><span style=\"font-weight: 400;\">Le d\u00e9veloppement de Stable Diffusion a impliqu\u00e9 des chercheurs d\u2019un groupe de l&rsquo;<strong>Universit\u00e9 Ludwig Maximilian<\/strong> et des donn\u00e9es d&rsquo;entra\u00eenement qui, \u00e0 en croire Mostaque, auraient \u00e9t\u00e9 fournies par des organisations \u00e0 but non lucratif. Nous verrons plus loin que cette affirmation a \u00e9t\u00e9 contest\u00e9e.<\/span><\/p><p><span style=\"font-weight: 400;\">Stable Diffusion n\u2019a pas \u00e9t\u00e9 la 1<\/span><span style=\"font-weight: 400;\">\u00e8re<\/span><span style=\"font-weight: 400;\"> IA g\u00e9n\u00e9rative d\u2019image propos\u00e9e au public. D\u00e8s janvier 2021, OpenAI a \u00e9t\u00e9 en mesure de pr\u00e9senter la premi\u00e8re application concr\u00e8te de ses recherches : Dall-e.\u00a0 Il a fallu attendre avril 2022 pour que Dall-e 2 apparaisse, et offre des capacit\u00e9s de g\u00e9n\u00e9ration d\u2019image abouties.<\/span><\/p><p><span style=\"font-weight: 400;\">Le 12 juillet 2022, une autre IA g\u00e9n\u00e9rative est entr\u00e9e sur le march\u00e9 (en version beta) : <a href=\"https:\/\/datascientest.com\/midjourney-tout-savoir\">Midjourney<\/a>. Elle a frapp\u00e9 les esprits par <strong>la qualit\u00e9 de ses cr\u00e9ations photo r\u00e9alistes<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Stable Diffusion, pour sa part, est apparu en ao\u00fbt 2022 et <strong>s\u2019est fait \u00e9galement remarquer par sa capacit\u00e9 \u00e0 g\u00e9n\u00e9rer des images sophistiqu\u00e9es<\/strong>. D\u00e8s le 17 octobre 2022, la startup a lev\u00e9 101 millions de dollars, \u00e9v\u00e9nement c\u00e9l\u00e9br\u00e9 lors d\u2019une soir\u00e9e m\u00e9morable \u00e0 San Francisco.<\/span><\/p><p><span style=\"font-weight: 400;\">Jusqu\u2019au milieu de l\u2019\u00e9t\u00e9 2023, Midjourney se montrait largement sup\u00e9rieur \u00e0 Stable Diffusion. A la toute fin juillet, Stable Diffusion XL 1.0 a vu le jour. Dans les jours qui ont suivi, les forums ont clam\u00e9 un m\u00eame message : le nouveau Stable Diffusion aurait \u00e9gal\u00e9 ou d\u00e9pass\u00e9 Midjourney !<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-0077024 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"0077024\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/prompt-engineer-tout-savoir\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Apprendre \u00e0 utiliser les IA g\u00e9n\u00e9ratives<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-8f633ec e-flex e-con-boxed e-con\" data-id=\"8f633ec\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-10bf0cb elementor-widget elementor-widget-heading\" data-id=\"10bf0cb\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Stable Diffusion XL<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-f568995 elementor-widget elementor-widget-text-editor\" data-id=\"f568995\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\"><a href=\"https:\/\/datascientest.com\/stable-diffusion-xl-tout-savoir\">Stable Diffusion XL<\/a> 1.0 (SDXL 1.0) a affirm\u00e9 son excellence dans un grand nombre de repr\u00e9sentations : cartoon, art num\u00e9rique, fantastique\u2026 Il se distingue par la qualit\u00e9 de contraste des couleurs, la gestion des ombres et lumi\u00e8res, la <strong>capacit\u00e9 \u00e0 produire des effets audacieux<\/strong>\u2026<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-ad26837 e-flex e-con-boxed e-con\" data-id=\"ad26837\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-b2498ab elementor-widget elementor-widget-heading\" data-id=\"b2498ab\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Comment acc\u00e9der \u00e0 Stable Diffusion ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-fb6b592 elementor-widget elementor-widget-text-editor\" data-id=\"fb6b592\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Stable Diffusion est accessible depuis un grand nombre de sites, notamment<\/span><\/p><ul><li><a href=\"https:\/\/clipdrop.co\/fr\"><span style=\"font-weight: 400;\">clipdrop.co<\/span><\/a><span style=\"font-weight: 400;\"> \u2013 accessible en fran\u00e7ais<\/span><\/li><li><a href=\"https:\/\/dreamstudio.ai\/\"><span style=\"font-weight: 400;\">dreamstudio.ai<\/span><\/a><\/li><li><a href=\"https:\/\/stablediffusionweb.com\/fr\"><span style=\"font-weight: 400;\">stablediffusionweb.com<\/span><\/a><\/li><li><span style=\"font-weight: 400;\">Night Cafe<\/span><\/li><li><span style=\"font-weight: 400;\">etc.<\/span><\/li><\/ul>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-fb8bf3e elementor-widget elementor-widget-text-editor\" data-id=\"fb8bf3e\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Si un grand nombre de formules sont gratuites, d\u2019autres n\u00e9cessitent d\u2019acheter des cr\u00e9dits \u00e0 un tarif raisonnable.<\/span><\/p><p><span style=\"font-weight: 400;\">De nombreux cr\u00e9atifs appr\u00e9cient le fait qu\u2019il soit possible d\u2019installer Stable Diffusion en version locale sur un ordinateur dot\u00e9 d\u2019une bonne carte graphique. De fait, son code, qui est open source, peut fonctionner sur la plupart des mat\u00e9riels \u00e9quip\u00e9s d&rsquo;un <a href=\"https:\/\/datascientest.com\/gpu-data-science\">GPU<\/a> avec au moins 4 Go de VRAM.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-698f910 elementor-widget elementor-widget-image\" data-id=\"698f910\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"875\" height=\"500\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion1.jpg\" class=\"attachment-full size-full wp-image-234629\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion1.jpg 875w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion1-300x171.jpg 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion1-768x439.jpg 768w\" sizes=\"(max-width: 875px) 100vw, 875px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-f3c522b e-flex e-con-boxed e-con\" data-id=\"f3c522b\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-b120e0f elementor-widget elementor-widget-heading\" data-id=\"b120e0f\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Les d\u00e9boires de Stability AI<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-fea821f elementor-widget elementor-widget-text-editor\" data-id=\"fea821f\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">H\u00e9las, l\u2019ann\u00e9e 2023 a \u00e9t\u00e9 difficile pour Stability AI qui a vu sa tr\u00e9sorerie fondre comme neige au soleil en raison des co\u00fbts d\u2019exploitation de Stable Diffusion et de nombreux recrutements. La startup londonienne a cherch\u00e9 \u00e0 lever 400 millions de dollars, mais en vain. Stability AI est par ailleurs cibl\u00e9 par deux proc\u00e9dures judiciaires. Des collectifs d&rsquo;artistes, mais aussi par Getty Images pr\u00e9tendent que Stabilit\u00e9 AI aurait utilis\u00e9 leurs \u0153uvres sans consentement pour entra\u00eener son d\u2019IA. La situation est devenue telle que l\u2019un de ses principaux investisseurs a demand\u00e9 la d\u00e9mission de Mostaque.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-ebefac5 e-flex e-con-boxed e-con\" data-id=\"ebefac5\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-ecc5caf elementor-widget elementor-widget-heading\" data-id=\"ecc5caf\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Stable Diffusion 3<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-88d241e elementor-widget elementor-widget-text-editor\" data-id=\"88d241e\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Stable Diffusion devrait toutefois rebondir avec sa version 3 annonc\u00e9e fin f\u00e9vrier 2024 et qui am\u00e9liore notablement la <strong>qualit\u00e9 de l&rsquo;image<\/strong>, et <strong>produire pr\u00e9cis\u00e9ment du texte dans une image<\/strong>, ce qui demeure un point faible de ses concurrents.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ce790a5 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"ce790a5\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formation-intelligence-artificielle\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Se former en intelligence artificielle<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-eeb3a56 elementor-widget elementor-widget-html\" data-id=\"eeb3a56\" data-element_type=\"widget\" data-widget_type=\"html.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<script type=\"application\/ld+json\">\r\n{\r\n  \"@context\": \"https:\/\/schema.org\",\r\n  \"@type\": \"FAQPage\",\r\n  \"mainEntity\": [{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Quelle est l'histoire de Stable Diffusion ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"Le d\u00e9veloppement de Stable Diffusion a impliqu\u00e9 des chercheurs d\u2019un groupe de l'Universit\u00e9 Ludwig Maximilian et des donn\u00e9es d'entra\u00eenement qui, \u00e0 en croire Mostaque, auraient \u00e9t\u00e9 fournies par des organisations \u00e0 but non lucratif. Nous verrons plus loin que cette affirmation a \u00e9t\u00e9 contest\u00e9e.\"\r\n    }\r\n  },{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Quels sont les points forts de Stable Diffusion XL ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"Stable Diffusion XL 1.0 (SDXL 1.0) a affirm\u00e9 son excellence dans un grand nombre de repr\u00e9sentations : cartoon, art num\u00e9rique, fantastique\u2026 Il se distingue par la qualit\u00e9 de contraste des couleurs, la gestion des ombres et lumi\u00e8res, la capacit\u00e9 \u00e0 produire des effets audacieux\u2026\"\r\n    }\r\n  },{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Qu'est-ce que Stable Diffusion 3 ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"Stable Diffusion devrait toutefois rebondir avec sa version 3 annonc\u00e9e fin f\u00e9vrier 2024 et qui am\u00e9liore notablement la qualit\u00e9 de l'image, et produire pr\u00e9cis\u00e9ment du texte dans une image, ce qui demeure un point faible de ses concurrents.\"\r\n    }\r\n  }]\r\n}\r\n<\/script>\r\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>La startup londonienne stability.ai a r\u00e9ussi l\u2019exploit d\u2019imposer une IA g\u00e9n\u00e9rative d\u2019images face \u00e0 deux mastodontes am\u00e9ricains. L\u2019application s\u2019appelle Stable Diffusion et elle a bien des atouts\u2026 Si l\u2019on questionne le grand public sur l\u2019intelligence artificielle (IA), deux noms surgissent spontan\u00e9ment. Le premier est bien \u00e9videmment ChatGPT, lequel domine tr\u00e8s largement le march\u00e9 de l\u2019IA. [&hellip;]<\/p>\n","protected":false},"author":78,"featured_media":234600,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"content-type":"","inline_featured_image":false},"categories":[2500],"tags":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.7 (Yoast SEO v21.7) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/datascientest.com\/stable-diffusion-tout-savoir\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datascientest.com\/stable-diffusion-tout-savoir\" \/>\n<meta property=\"og:site_name\" content=\"Formation Data Science | DataScientest.com\" \/>\n<meta property=\"article:published_time\" content=\"2024-04-09T06:30:00+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-04-03T09:25:05+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"600\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"J\u00e9r\u00e9my Robert\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"J\u00e9r\u00e9my Robert\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"6 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/datascientest.com\/stable-diffusion-tout-savoir\",\"url\":\"https:\/\/datascientest.com\/stable-diffusion-tout-savoir\",\"name\":\"Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/#website\"},\"datePublished\":\"2024-04-09T06:30:00+00:00\",\"dateModified\":\"2024-04-03T09:25:05+00:00\",\"author\":{\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/datascientest.com\/stable-diffusion-tout-savoir\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/datascientest.com\/#website\",\"url\":\"https:\/\/datascientest.com\/\",\"name\":\"Formation Data Science | DataScientest.com\",\"description\":\"Leader de la formation en Data Science\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/datascientest.com\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925\",\"name\":\"J\u00e9r\u00e9my Robert\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg\",\"contentUrl\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg\",\"caption\":\"J\u00e9r\u00e9my Robert\"},\"description\":\"J\u00e9r\u00e9my Robert est actuellement Directeur p\u00e9dagogique B2B chez DataScientest.com. Avant de rejoindre DataScientest, il a travaill\u00e9 en tant que Consultant Business Intelligence chez DEAPDATA, o\u00f9 il a g\u00e9r\u00e9 des projets dans les domaines de l'assurance, bancaire et industrie. Il a \u00e9galement \u00e9t\u00e9 Co-fondateur et pr\u00e9sident de Tripzer. En tant que consultant ind\u00e9pendant, J\u00e9r\u00e9my a propos\u00e9 des formations sur MicroStrategy pour divers niveaux professionnels. Il est dipl\u00f4m\u00e9 de l'Universit\u00e9 Paris Cit\u00e9 avec un Master MIAGE, combinant comp\u00e9tences en informatique et gestion.\",\"sameAs\":[\"https:\/\/www.linkedin.com\/in\/jeremyrobert1\/\",\"Data Science\"],\"url\":\"https:\/\/datascientest.com\/author\/jeremyrobert\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/datascientest.com\/stable-diffusion-tout-savoir","og_locale":"fr_FR","og_type":"article","og_title":"Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning","og_url":"https:\/\/datascientest.com\/stable-diffusion-tout-savoir","og_site_name":"Formation Data Science | DataScientest.com","article_published_time":"2024-04-09T06:30:00+00:00","article_modified_time":"2024-04-03T09:25:05+00:00","og_image":[{"width":1200,"height":600,"url":"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/03\/stable_diffusion.png","type":"image\/png"}],"author":"J\u00e9r\u00e9my Robert","twitter_card":"summary_large_image","twitter_misc":{"\u00c9crit par":"J\u00e9r\u00e9my Robert","Dur\u00e9e de lecture estim\u00e9e":"6 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/datascientest.com\/stable-diffusion-tout-savoir","url":"https:\/\/datascientest.com\/stable-diffusion-tout-savoir","name":"Stable Diffusion : tout savoir sur ce mod\u00e8le de Machine Learning","isPartOf":{"@id":"https:\/\/datascientest.com\/#website"},"datePublished":"2024-04-09T06:30:00+00:00","dateModified":"2024-04-03T09:25:05+00:00","author":{"@id":"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datascientest.com\/stable-diffusion-tout-savoir"]}]},{"@type":"WebSite","@id":"https:\/\/datascientest.com\/#website","url":"https:\/\/datascientest.com\/","name":"Formation Data Science | DataScientest.com","description":"Leader de la formation en Data Science","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datascientest.com\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"fr-FR"},{"@type":"Person","@id":"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925","name":"J\u00e9r\u00e9my Robert","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/datascientest.com\/#\/schema\/person\/image\/","url":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg","contentUrl":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg","caption":"J\u00e9r\u00e9my Robert"},"description":"J\u00e9r\u00e9my Robert est actuellement Directeur p\u00e9dagogique B2B chez DataScientest.com. Avant de rejoindre DataScientest, il a travaill\u00e9 en tant que Consultant Business Intelligence chez DEAPDATA, o\u00f9 il a g\u00e9r\u00e9 des projets dans les domaines de l'assurance, bancaire et industrie. Il a \u00e9galement \u00e9t\u00e9 Co-fondateur et pr\u00e9sident de Tripzer. En tant que consultant ind\u00e9pendant, J\u00e9r\u00e9my a propos\u00e9 des formations sur MicroStrategy pour divers niveaux professionnels. Il est dipl\u00f4m\u00e9 de l'Universit\u00e9 Paris Cit\u00e9 avec un Master MIAGE, combinant comp\u00e9tences en informatique et gestion.","sameAs":["https:\/\/www.linkedin.com\/in\/jeremyrobert1\/","Data Science"],"url":"https:\/\/datascientest.com\/author\/jeremyrobert"}]}},"_links":{"self":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/234598"}],"collection":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/users\/78"}],"replies":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/comments?post=234598"}],"version-history":[{"count":0,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/234598\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media\/234600"}],"wp:attachment":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media?parent=234598"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/categories?post=234598"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/tags?post=234598"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}