1:"$Sreact.fragment"
3:I[54118,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1737","static/chunks/fe207bbb-c84ed91deca4f888.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","144","static/chunks/144-1a048554a0613a1b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","528","static/chunks/528-7ef2006e50a52549.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2530","static/chunks/2530-9eb218eca5ffc00b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7177","static/chunks/app/layout-75dbb92d3cda2f1c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"Providers"]
4:I[98195,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1737","static/chunks/fe207bbb-c84ed91deca4f888.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","144","static/chunks/144-1a048554a0613a1b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","528","static/chunks/528-7ef2006e50a52549.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2530","static/chunks/2530-9eb218eca5ffc00b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7177","static/chunks/app/layout-75dbb92d3cda2f1c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"GoogleTagManager"]
5:I[4411,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1737","static/chunks/fe207bbb-c84ed91deca4f888.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","144","static/chunks/144-1a048554a0613a1b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","528","static/chunks/528-7ef2006e50a52549.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2530","static/chunks/2530-9eb218eca5ffc00b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7177","static/chunks/app/layout-75dbb92d3cda2f1c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"FilmAwardStrip"]
6:"$Sreact.suspense"
7:I[40459,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1737","static/chunks/fe207bbb-c84ed91deca4f888.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","144","static/chunks/144-1a048554a0613a1b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","528","static/chunks/528-7ef2006e50a52549.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2530","static/chunks/2530-9eb218eca5ffc00b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7177","static/chunks/app/layout-75dbb92d3cda2f1c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"Navbar"]
8:I[15848,[],""]
9:I[55256,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","8039","static/chunks/app/error-332e154bd557e323.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"default"]
a:I[62142,[],""]
19:I[79302,["4219","static/chunks/app/global-error-f07d85f704128170.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"default"]
:HL["/_next/static/media/7d4881bb7e1bf84d-s.p.woff2","font",{"crossOrigin":"","type":"font/woff2"}]
:HL["/_next/static/css/c91c557421edbcc1.css?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","style"]
:HL["/_next/static/css/9c4f0c3be44ead33.css?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","style"]
2:T412,
(function () {
  window.dataLayer = window.dataLayer || [];
  function gtag() { window.dataLayer.push(arguments); }

  gtag('consent', 'default', {
    ad_storage: 'denied',
    ad_user_data: 'denied',
    ad_personalization: 'denied',
    analytics_storage: 'denied',
    functionality_storage: 'denied',
    personalization_storage: 'denied',
    security_storage: 'granted'
  });

  try {
    var match = document.cookie.match(/(?:^|;\s*)gdpr_consent=([^;]*)/);
    if (!match) return;
    var saved = JSON.parse(decodeURIComponent(match[1]));
    if (!saved || typeof saved !== 'object') return;
    gtag('consent', 'update', {
      analytics_storage: saved.analytics ? 'granted' : 'denied',
      ad_storage: saved.marketing ? 'granted' : 'denied',
      ad_user_data: saved.marketing ? 'granted' : 'denied',
      ad_personalization: saved.marketing ? 'granted' : 'denied',
      functionality_storage: saved.functional ? 'granted' : 'denied'
    });
  } catch (e) {
    /* malformed cookie â€” the denied default stands */
  }
})();
0:{"P":null,"b":"vX-6P7rGIK5v7u82yfIum","p":"","c":["","models","gpt-image-2"],"i":false,"f":[[["",{"children":["models",{"children":[["route","gpt-image-2","d"],{"children":["__PAGE__",{}]}]}]},"$undefined","$undefined",true],["",["$","$1","c",{"children":[[["$","link","0",{"rel":"stylesheet","href":"/_next/static/css/c91c557421edbcc1.css?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","precedence":"next","crossOrigin":"$undefined","nonce":"$undefined"}],["$","link","1",{"rel":"stylesheet","href":"/_next/static/css/9c4f0c3be44ead33.css?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","precedence":"next","crossOrigin":"$undefined","nonce":"$undefined"}]],["$","html",null,{"lang":"en","suppressHydrationWarning":true,"className":"__variable_8c25df","children":[["$","head",null,{"children":[["$","script",null,{"id":"consent-mode-default","dangerouslySetInnerHTML":{"__html":"$2"}}],["$","script",null,{"id":"film-award-dismissed","dangerouslySetInnerHTML":{"__html":"try{if(document.cookie.split(';').some(function(c){return c.trim()===\"film_award_2027_dismissed=1\"}))document.documentElement.setAttribute(\"data-film-award-dismissed\",'')}catch(e){}"}}],["$","link",null,{"rel":"preconnect","href":"https://images.segmind.com"}],["$","link",null,{"rel":"preconnect","href":"https://segmind-resources.s3.amazonaws.com"}],["$","link",null,{"rel":"preconnect","href":"https://www.googletagmanager.com"}],["$","link",null,{"rel":"preload","href":"/fonts/Satoshi-Variable.woff2","as":"font","type":"font/woff2","crossOrigin":"anonymous"}]]}],["$","body",null,{"className":"antialiased bg-background text-foreground","children":["$","$L3",null,{"children":[["$","$L4",null,{"gtmId":"GTM-MHW23H4Z"}],["$","div",null,{"className":"min-h-screen flex flex-col","children":[["$","$L5",null,{}],["$","$6",null,{"fallback":["$","div",null,{"className":"h-16 bg-background/95 backdrop-blur border-b border-border"}],"children":["$","$L7",null,{}]}],["$","main",null,{"className":"flex-1","children":["$","$L8",null,{"parallelRouterKey":"children","error":"$9","errorStyles":[],"errorScripts":[],"template":["$","$La",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","div",null,{"className":"min-h-screen bg-background flex items-center justify-center px-4 sm:px-6 lg:px-8","children":["$","div",null,{"className":"max-w-lg w-full text-center","children":[["$","div",null,{"className":"mb-8","children":["$","div",null,{"className":"relative","children":[["$","h1",null,{"className":"text-9xl font-bold text-muted/20 dark:text-muted/10 select-none","children":"404"}],["$","div",null,{"className":"absolute inset-0 flex items-center justify-center","children":["$","div",null,{"className":"relative","children":[["$","div",null,{"className":"absolute -top-8 -left-8 w-16 h-16 bg-brand-100 dark:bg-brand-950 rounded-full flex items-center justify-center animate-bounce","children":["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","stroke":"currentColor","strokeWidth":2,"strokeLinecap":"round","strokeLinejoin":"round","className":"lucide lucide-zap w-8 h-8 text-brand-500","children":["$Lb","$undefined"]}]}],"$Lc","$Ld"]}]}]]}]}],"$Le","$Lf","$L10","$L11"]}]}],[]],"forbidden":"$undefined","unauthorized":"$undefined"}]}],"$L12"]}],"$L13","$L14"]}]}]]}]]}],{"children":["models","$L15",{"children":[["route","gpt-image-2","d"],"$L16",{"children":["__PAGE__","$L17",{},null,false]},null,false]},null,false]},null,false],"$L18",false]],"m":"$undefined","G":["$19",[]],"s":false,"S":true}
1a:I[75289,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6541","static/chunks/fb4e7fc9-e31ad622f5fde25d.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9761","static/chunks/998e86a0-e112630ec4db17dc.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6118","static/chunks/6118-7328b7d3f629a5ef.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","393","static/chunks/393-7cd2157fc3790475.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3985","static/chunks/3985-1f9bcb95de313d61.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4844","static/chunks/4844-011fe3465334e7d0.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9305","static/chunks/9305-0f5d0f2a661afddc.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","713","static/chunks/713-dcb3ca4f0233704d.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3504","static/chunks/3504-ae9bc2dc93d2ef47.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1885","static/chunks/1885-f206f5873f09b3e2.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5126","static/chunks/5126-bdde68ce894e4935.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5642","static/chunks/5642-c80cb7ee6e914a6a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6372","static/chunks/6372-3e2880fc7d769d1f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1289","static/chunks/1289-4fdf72847f7ef615.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","8382","static/chunks/8382-004a14550e3769cc.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1951","static/chunks/app/models/%5Broute%5D/page-c5fd4df0a8a5cab4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],""]
1b:I[4717,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1737","static/chunks/fe207bbb-c84ed91deca4f888.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","144","static/chunks/144-1a048554a0613a1b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","528","static/chunks/528-7ef2006e50a52549.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2530","static/chunks/2530-9eb218eca5ffc00b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7177","static/chunks/app/layout-75dbb92d3cda2f1c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"Footer"]
1c:I[35733,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","341","static/chunks/app/models/(catalogue)/loading-03c2017343ab8539.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"Toaster"]
1d:I[20659,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1737","static/chunks/fe207bbb-c84ed91deca4f888.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6397","static/chunks/6397-b7ccd1edf8aedf12.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","144","static/chunks/144-1a048554a0613a1b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","528","static/chunks/528-7ef2006e50a52549.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2530","static/chunks/2530-9eb218eca5ffc00b.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7177","static/chunks/app/layout-75dbb92d3cda2f1c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"LazyLayoutExtras"]
20:I[30661,[],"OutletBoundary"]
22:I[72272,[],"AsyncMetadataOutlet"]
24:I[30661,[],"ViewportBoundary"]
26:I[30661,[],"MetadataBoundary"]
b:["$","path","1xq2db",{"d":"M4 14a1 1 0 0 1-.78-1.63l9.9-10.2a.5.5 0 0 1 .86.46l-1.92 6.02A1 1 0 0 0 13 10h7a1 1 0 0 1 .78 1.63l-9.9 10.2a.5.5 0 0 1-.86-.46l1.92-6.02A1 1 0 0 0 11 14z"}]
c:["$","div",null,{"className":"absolute -top-4 -right-12 w-12 h-12 bg-brand-secondary-100 dark:bg-brand-secondary-950 rounded-full flex items-center justify-center animate-pulse","children":["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","stroke":"currentColor","strokeWidth":2,"strokeLinecap":"round","strokeLinejoin":"round","className":"lucide lucide-search w-6 h-6 text-brand-secondary-500","children":[["$","circle","4ej97u",{"cx":"11","cy":"11","r":"8"}],["$","path","1qie3q",{"d":"m21 21-4.3-4.3"}],"$undefined"]}]}]
d:["$","div",null,{"className":"absolute -bottom-6 left-4 w-10 h-10 bg-green-100 dark:bg-green-950 rounded-full flex items-center justify-center animate-bounce delay-200","children":["$","div",null,{"className":"w-3 h-3 bg-green-500 rounded-full"}]}]
e:["$","div",null,{"className":"mb-8","children":[["$","h2",null,{"className":"text-2xl font-semibold text-foreground mb-4","children":"Page Not Found"}],["$","p",null,{"className":"text-muted-foreground text-lg mb-2","children":"Oops! The page you're looking for doesn't exist."}],["$","p",null,{"className":"text-muted-foreground","children":"It might have been moved, deleted, or you entered the wrong URL."}]]}]
f:["$","div",null,{"className":"flex flex-col sm:flex-row gap-4 justify-center items-center mb-12","children":[["$","$L1a",null,{"href":"/","className":"inline-flex items-center justify-center gap-2 px-6 py-3 bg-brand-600 dark:bg-brand-500 text-white rounded-lg hover:bg-brand-700 dark:hover:bg-brand-600 transition-colors font-medium text-base w-full sm:w-auto","children":[["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","stroke":"currentColor","strokeWidth":2,"strokeLinecap":"round","strokeLinejoin":"round","className":"lucide lucide-house w-5 h-5","children":[["$","path","5wwlr5",{"d":"M15 21v-8a1 1 0 0 0-1-1h-4a1 1 0 0 0-1 1v8"}],["$","path","1d0kgt",{"d":"M3 10a2 2 0 0 1 .709-1.528l7-5.999a2 2 0 0 1 2.582 0l7 5.999A2 2 0 0 1 21 10v9a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2z"}],"$undefined"]}],"Go Home"]}],["$","$L1a",null,{"href":"/models","className":"inline-flex items-center justify-center gap-2 px-6 py-3 bg-surface border border-border text-foreground rounded-lg hover:bg-surface/80 transition-colors font-medium text-base w-full sm:w-auto","children":[["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","stroke":"currentColor","strokeWidth":2,"strokeLinecap":"round","strokeLinejoin":"round","className":"lucide lucide-search w-5 h-5","children":[["$","circle","4ej97u",{"cx":"11","cy":"11","r":"8"}],["$","path","1qie3q",{"d":"m21 21-4.3-4.3"}],"$undefined"]}],"Browse Models"]}]]}]
10:["$","div",null,{"className":"border-t border-border pt-8","children":[["$","p",null,{"className":"text-sm text-muted-foreground mb-4","children":"Popular destinations:"}],["$","div",null,{"className":"flex flex-wrap justify-center gap-4 text-sm","children":[["$","$L1a",null,{"href":"/models/tags/image","className":"text-brand-600 dark:text-brand-400 hover:text-brand-700 dark:hover:text-brand-300 transition-colors","children":"Image Models"}],["$","span",null,{"className":"text-muted-foreground","children":"â€¢"}],["$","$L1a",null,{"href":"/models/tags/video","className":"text-brand-600 dark:text-brand-400 hover:text-brand-700 dark:hover:text-brand-300 transition-colors","children":"Video Models"}],["$","span",null,{"className":"text-muted-foreground","children":"â€¢"}],["$","$L1a",null,{"href":"/models/tags/llms","className":"text-brand-600 dark:text-brand-400 hover:text-brand-700 dark:hover:text-brand-300 transition-colors","children":"LLMs"}],["$","span",null,{"className":"text-muted-foreground","children":"â€¢"}],["$","$L1a",null,{"href":"/models","className":"text-brand-600 dark:text-brand-400 hover:text-brand-700 dark:hover:text-brand-300 transition-colors","children":"All Models"}]]}]]}]
11:["$","div",null,{"className":"mt-8","children":["$","$L1a",null,{"href":"/","className":"inline-flex items-center text-sm text-muted-foreground hover:text-foreground transition-colors","children":[["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","stroke":"currentColor","strokeWidth":2,"strokeLinecap":"round","strokeLinejoin":"round","className":"lucide lucide-arrow-left w-4 h-4 mr-2","children":[["$","path","1l729n",{"d":"m12 19-7-7 7-7"}],["$","path","x3x0zl",{"d":"M19 12H5"}],"$undefined"]}],"Go back to homepage"]}]}]
12:["$","$L1b",null,{}]
13:["$","$L1c",null,{"closeButton":true,"toastOptions":{"duration":30000}}]
14:["$","$L1d",null,{}]
15:["$","$1","c",{"children":[null,["$","$L8",null,{"parallelRouterKey":"children","error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$La",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":"$undefined","forbidden":"$undefined","unauthorized":"$undefined"}]]}]
16:["$","$1","c",{"children":[null,"$L1e"]}]
17:["$","$1","c",{"children":["$L1f",null,["$","$L20",null,{"children":["$L21",["$","$L22",null,{"promise":"$@23"}]]}]]}]
18:["$","$1","h",{"children":[null,[["$","$L24",null,{"children":"$L25"}],["$","meta",null,{"name":"next-size-adjust","content":""}]],["$","$L26",null,{"children":["$","div",null,{"hidden":true,"children":["$","$6",null,{"fallback":null,"children":"$L27"}]}]}]]}]
25:[["$","meta","0",{"charSet":"utf-8"}],["$","meta","1",{"name":"viewport","content":"width=device-width, initial-scale=1"}]]
21:null
28:I[53921,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7213","static/chunks/7213-ed3acee0c865cd60.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7256","static/chunks/app/models/%5Broute%5D/layout-23522cb6540a765c.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"default"]
2d:I[4704,[],"IconMark"]
29:Te48,# GPT Image 2: Photorealistic Text-to-Image and Edit Model

## What is GPT Image 2?

GPT Image 2 is OpenAI's next-generation image model, launched in April 2026 as the successor to gpt-image-1.5. It generates photorealistic images from text or edits existing images guided by a prompt, all through a single endpoint. The headline improvement is near-perfect in-image typography: over 95% accuracy across Latin, Japanese, Korean, Chinese, Hindi Devanagari, and Bengali scripts â€” the first image model practical for shipping UI labels, posters, and multilingual marketing assets without a manual redraw pass. A new single-pass architecture roughly doubles generation speed over the previous version, and built-in reasoning plans composition, counts items, and checks constraints before rendering.

## Key Features

- Text-to-image generation and guided image editing in one API
- In-image text rendering at 95%+ accuracy, including non-Latin scripts
- Output resolutions up to 2K across landscape (`1536x1024`), portrait (`1024x1536`), and square (`1024x1024`)
- Output formats: PNG (sharpest text), WebP (smaller files), JPEG (universal)
- Moderation controls: `auto` default, `low` for permitted use cases
- Native multi-constraint prompt adherence at ~98% accuracy

## Best Use Cases

GPT Image 2 is the right choice anytime legible text is part of the image: magazine covers with headlines, product packaging mockups, storefront and signboard scenes, infographics and charts, storyboards and comic panels, multilingual ad creatives, UI screen mockups, and posters. In testing, it rendered a handwritten chalkboard easel combining English ("Mumbai Book Store", "Open Daily 9 am â€“ 9 pm") and Hindi Devanagari ("à¤®à¥à¤‚à¤¬à¤ˆ à¤ªà¥à¤¸à¥à¤¤à¤• à¤­à¤‚à¤¡à¤¾à¤°") cleanly on the first try. Edit mode (passing an `image` input) is ideal for relighting, background swaps, text changes on existing visuals, and brand-consistent variations.

## Prompt Tips and Output Quality

Keep `quality=high` whenever typography matters â€” `medium` and `low` degrade fine lettering. Lead the prompt with subject, then typography in quotes, then style and lighting cues. For magazine-style layouts pick `1024x1536`; for marketing banners and scenes, `1536x1024`. Keep `output_format=png` and `output_compression=100` when text crispness is non-negotiable.

## FAQs

**Does GPT Image 2 render text in Hindi, Japanese, and Chinese?**
Yes. Multilingual typography is the model's flagship capability â€” Devanagari, CJK, Korean, and Bengali all render cleanly enough to ship.

**What is the difference between generation and edit mode?**
Leaving the `image` parameter null generates from text alone. Passing an image URL switches the model into edit mode, where the prompt guides modifications to the input.

**What output sizes are supported?**
`1024x1024`, `1536x1024`, `1024x1536`, and `auto`. All run up to 2K resolution with high quality.

**Does GPT Image 2 support transparent backgrounds?**
No. `background` accepts `opaque` (the default) or `auto`; requesting `transparent` returns a 400. Use `gpt-image-1.5` when you need an alpha channel for logos, stickers, or product cutouts.

**Is GPT Image 2 faster than gpt-image-1.5?**
Yes â€” roughly 2Ã— faster thanks to a new single-pass architecture, with fewer artifacts on hands, faces, and material surfaces.

**Where does GPT Image 2 fall short?**
Physical reasoning tasks (origami, angled reflections, Rubik's cubes) and highly dense repetitive detail (circuit diagrams, grains of sand) still challenge the model. Iterative edits beyond one or two passes tend to drift.
2a:T951,# How to Use GPT Image 2

GPT Image 2 handles both text-to-image generation and guided editing through a single endpoint. The way you configure its parameters decides whether output lands at magazine-ready quality or drifts into preview territory â€” especially when in-image typography is part of the brief.

## Structuring the Prompt

Lead with the subject and scene, then the exact text you want rendered (in quotes), then style and lighting cues. Specify fonts, scripts, and hierarchy when shipping multilingual creatives. Example: *"A chalkboard easel reads 'Open Daily 9 am â€“ 9 pm' in clean white chalk, shallow depth of field, golden-hour lighting."* The model handles Latin, Japanese, Korean, Chinese, Hindi Devanagari, and Bengali without script switching tricks.

## Recommended Settings

Based on working test runs, these defaults produce production-ready output:

- **`quality=high`** â€” the single most important setting. Drop it only for quick previews; `medium` and `low` visibly degrade lettering.
- **`size`** â€” `1536x1024` for marketing banners and scene compositions; `1024x1536` for magazine covers, posters, and portrait social; `1024x1024` for in-feed social.
- **`output_format=png`** with `output_compression=100` whenever text crispness matters. Switch to `webp` for web delivery where file size is a priority.
- **`moderation=auto`** is the safe default; switch to `low` only for permitted use cases.
- **`background=opaque`** (the default) or `auto` â€” this model does not support transparent output. For cutouts that need an alpha channel, use `gpt-image-1.5`.

## Text-to-Image vs Edit Mode

Leave the `image` parameter null for pure text-to-image generation. To edit an existing image â€” relighting, background swap, on-image text change, brand variations â€” pass the image URL and the model switches into edit mode automatically. The same prompt guidelines apply: be specific about what you want changed and what must be preserved.

## Best Practices

Batch multi-constraint prompts (object counts, colors, text strings, positions) in a single pass â€” GPT Image 2 resolves around 98% of these without rerolls. Avoid iterative editing beyond one or two passes; quality plateaus and drift creeps in. For dense technical diagrams or physically constrained geometry (origami, reflections), validate outputs before shipping.
2b:T8ae,## How Pricing Works

GPT Image 2 is billed per token: $8.00 per million input tokens (your prompt and any reference image) and $30.00 per million output tokens (the generated image). Output cost dominates in almost every real workload, so the `quality` parameter and the output `size` are the two levers that actually move the bill. All other parameters (`output_format`, `output_compression`, `background`, `moderation`) are free.

### Pricing Examples

Costs below are per call, measured from live API header deductions on April 22, 2026.

| Size | Low Quality | Medium Quality | High Quality |
|---|---|---|---|
| 1024 x 768 | $0.01 | $0.04 | $0.15 |
| 1024 x 1024 | $0.01 | $0.06 | $0.22 |
| 1024 x 1536 | $0.01 | $0.05 | $0.17 |
| 1920 x 1080 | $0.01 | $0.04 | $0.16 |
| 2560 x 1440 | $0.01 | $0.06 | $0.23 |
| 3840 x 2160 | $0.02 | $0.11 | $0.41 |

### Cost Optimization Tips

- Use `quality=low` or `quality=medium` for ideation passes. At `low` quality every supported size is $0.01â€“$0.02 per image, so you can burn through 15â€“20 exploratory prompts for the cost of one `high` 4K render.
- `1024x768`, `1024x1024`, and `1920x1080` are the cheapest production sizes at `high` quality ($0.15â€“$0.22). Reach for `2560x1440` or `3840x2160` only when delivery resolution actually demands it â€” 4K `high` is nearly 3x the cost of 1080p `high`.
- `output_format` and `output_compression` do not change the price. Pick `jpeg` at compression 90 for CDN delivery, `png` at 100 when text crispness is non-negotiable.
- Edit mode (passing an `image` input) adds input tokens for the reference. Keep reference images under 1024px on the longest edge unless you need the detail.

### Example Scenario

A marketing agency shipping 20 localized Instagram Story ad variants per week at `1024x1536` and `quality=high` pays roughly $0.17 per image, or $3.40 per week. A production house generating 500 concert poster drafts per month at `quality=medium` (1024x1024) before selecting finalists to re-render at `high` lands near $40 per month, all-in. Teams delivering 4K hero assets at `3840x2160` `quality=high` should budget $0.41 per image â€” roughly 2.4x the cost of the same shot at 1024x1536.
1e:["$","$L28",null,{"route":"gpt-image-2","modelDetails":{"title":"GPT Image 2","thumbnail_image":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-3736.webp","type":"textToImage","model_type":"SO","is_depreciated":false,"meta_description":"Generate and edit photorealistic images with crisp multilingual typography, 2K resolution, and ~98% prompt adherence from a single API.","seo_description":"$29","guide":"$2a","pricing_guide":"$2b","artifacts":[],"default_image_output":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-example-output.png","default_response":null,"parameters":{"size":{"type":"str","label":"Size","hidden":false,"isEnum":true,"values":["1024x1024","1536x1024","1024x1536","1280x960","960x1280","1536x864","864x1536","2048x2048","2048x1152","1152x2048","3840x2160","2160x3840","auto"],"default":"1536x1024","component":"dropdown","isAdvanced":false,"isRequired":false,"description":"Output resolution (WIDTHxHEIGHT). 'auto' lets the model pick. For a custom resolution not listed, set Width and Height instead. Constraints: each edge a multiple of 16, aspect ratio 1:3-3:1, longest edge <= 3840, total pixels 655,360-8,294,400.","displayOrder":4,"displayValue":"1536x1024","affectsPricing":true},"width":{"type":"int","label":"Width","hidden":false,"isEnum":false,"default":null,"component":"input","isAdvanced":true,"isRequired":false,"description":"Optional custom width in pixels. Set both Width and Height to override the Size preset. Rules: each a multiple of 16, aspect ratio between 1:3 and 3:1, longest edge <= 3840, total pixels 655,360-8,294,400.","displayOrder":12,"displayValue":null,"affectsPricing":true},"height":{"type":"int","label":"Height","hidden":false,"isEnum":false,"default":null,"component":"input","isAdvanced":true,"isRequired":false,"description":"Optional custom height in pixels. Set both Width and Height to override the Size preset. Rules: each a multiple of 16, aspect ratio between 1:3 and 3:1, longest edge <= 3840, total pixels 655,360-8,294,400.","displayOrder":13,"displayValue":null,"affectsPricing":true},"prompt":{"type":"str","label":"Prompt","hidden":false,"isEnum":false,"component":"textarea","isAdvanced":false,"isRequired":true,"description":"Text describing the image; supports in-image typography across scripts. Lead with subject, style, lighting.","displayOrder":1,"displayValue":"A photorealistic, cinematic shot of a cozy independent bookstore in Mumbai at golden hour. Warm afternoon sunlight streams through a tall front window onto wooden shelves packed with books; book spines are clearly visible with titles in English and Hindi Devanagari script. In the foreground, a handwritten chalkboard A-frame easel reads, in clean legible chalk lettering: first line 'à¤®à¥à¤‚à¤¬à¤ˆ à¤ªà¥à¤¸à¥à¤¤à¤• à¤­à¤‚à¤¡à¤¾à¤°', second line 'Mumbai Book Store', and a smaller third line 'Open Daily  9 am â€“ 9 pm'. Shallow depth of field, shot on a full-frame camera at 35mm f/2.0, ultra-realistic detail, natural color grading, dust motes in sunlight, film grain.","affectsPricing":true},"quality":{"type":"str","label":"Quality","hidden":false,"isEnum":true,"values":["low","medium","high","auto"],"default":"high","component":"dropdown","isAdvanced":true,"isRequired":false,"description":"Rendering fidelity; 'high' keeps typography crisp. Use 'medium' or 'low' only for previews.","displayOrder":5,"displayValue":"high","affectsPricing":true},"background":{"type":"str","label":"Background","hidden":false,"isEnum":true,"values":["opaque"],"default":"opaque","component":"dropdown","isAdvanced":true,"isRequired":false,"description":"'opaque' for full scenes; 'transparent' for logos, stickers, and product cutouts.","displayOrder":7,"displayValue":"opaque","affectsPricing":false},"image_urls":{"type":"str","label":"Input Image URLs","format":"uri","hidden":false,"isEnum":false,"default":[],"component":"textbox","isAdvanced":false,"isRequired":false,"description":"A list of reference images. Include one or more URLs to edit or draw context from.","displayOrder":2,"displayValue":"","affectsPricing":true},"moderation":{"type":"str","label":"Moderation","hidden":false,"isEnum":true,"values":["low","auto"],"default":"auto","component":"dropdown","isAdvanced":true,"isRequired":false,"description":"Content filter strictness; 'auto' is the safe default. Use 'low' only for permitted use cases.","displayOrder":6,"displayValue":"auto","affectsPricing":false},"output_format":{"type":"str","label":"Output Format","hidden":false,"isEnum":true,"values":["png","jpeg","webp"],"default":"png","component":"dropdown","isAdvanced":true,"isRequired":false,"description":"Use 'png' for crisp text, 'webp' for size, 'jpeg' for broad compatibility.","displayOrder":9,"displayValue":"png","affectsPricing":false},"mask_image_url":{"type":"str","label":"Mask Image URL","format":"uri","hidden":false,"isEnum":false,"default":"","component":"textbox","isAdvanced":false,"isRequired":false,"description":"Optional mask image URL for surgical inpainting. White regions of the mask indicate areas to edit; everything outside stays pixel-perfect.","displayOrder":3,"displayValue":"","affectsPricing":false},"output_compression":{"type":"int","label":"Output Compression","hidden":false,"isEnum":false,"default":100,"component":"input","isAdvanced":true,"isRequired":false,"description":"Compression level 0-100; 100 preserves text crispness. Lower values reduce file size.","displayOrder":8,"displayValue":100,"affectsPricing":false}},"endpoints":{"main":"/v1/gpt-image-2"},"api_notes":[],"model_outputs":null,"llm_capabilities":null,"license":"","github_link":null,"inference_provider":"OPENAI","created_ts":"2026-04-22 05:04:46.746223+00:00","slug":"gpt-image-2","name":"GPT Image 2","description":"Generate photorealistic images with legible multilingual text and 2K output.","category":"textToImage","modality":"image","input":"text","output":"image","provider":"OPENAI","thumbnail_url":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-3736.webp","is_new":false,"coming_soon":false,"popularity_score":91,"category_percentile":99,"average_cost":0.039338914469471985,"latency_seconds":23.56594,"created_at":"2026-04-22 05:04:46.746223+00:00","parameters_schema":{"$schema":"https://json-schema.org/draft/2020-12/schema","type":"object","properties":{"prompt":{"type":"string","title":"Prompt","description":"Text describing the image. Supports in-image typography across scripts â€” lead with subject, style, lighting.","default":"A photorealistic, cinematic shot of a cozy independent bookstore in Mumbai at golden hour. Warm afternoon sunlight streams through a tall front window onto wooden shelves packed with books; book spines are clearly visible with titles in English and Hindi Devanagari script. In the foreground, a handwritten chalkboard A-frame easel reads, in clean legible chalk lettering: first line 'à¤®à¥à¤‚à¤¬à¤ˆ à¤ªà¥à¤¸à¥à¤¤à¤• à¤­à¤‚à¤¡à¤¾à¤°', second line 'Mumbai Book Store', and a smaller third line 'Open Daily  9 am â€“ 9 pm'. Shallow depth of field, shot on a full-frame camera at 35mm f/2.0, ultra-realistic detail, natural color grading, dust motes in sunlight, film grain.","minLength":1,"x-ui-metadata":{"order":0,"advanced":false,"hidden":false,"placeholder":"Describe the image you want to create"}},"image_urls":{"type":"array","title":"Reference images","description":"Optional reference images to edit or draw context from. Providing any image switches to edit mode.","default":[],"items":{"type":"string","format":"uri","x-file-metadata":{"allowed-types":["image/*"]}},"x-ui-metadata":{"order":1,"advanced":false,"hidden":false}},"size":{"type":"string","title":"Size","description":"Output resolution (WIDTHxHEIGHT). \"Auto\" lets the model pick.","default":"1536x1024","oneOf":[{"const":"1024x1024","title":"1024x1024"},{"const":"1536x1024","title":"1536x1024"},{"const":"1024x1536","title":"1024x1536"},{"const":"1280x960","title":"1280x960"},{"const":"960x1280","title":"960x1280"},{"const":"1536x864","title":"1536x864"},{"const":"864x1536","title":"864x1536"},{"const":"2048x2048","title":"2048x2048"},{"const":"2048x1152","title":"2048x1152"},{"const":"1152x2048","title":"1152x2048"},{"const":"3840x2160","title":"3840x2160"},{"const":"2160x3840","title":"2160x3840"},{"const":"auto","title":"Auto"}],"x-ui-metadata":{"order":2,"advanced":false,"hidden":false}},"mask_image_url":{"type":"string","title":"Mask image","description":"Optional mask for surgical inpainting, used only alongside a reference image. Must be a PNG with an alpha channel, matching the reference image's dimensions: the transparent areas are regenerated, everything opaque stays pixel-perfect. A flat black-and-white mask with no alpha channel is rejected.","format":"uri","x-file-metadata":{"allowed-types":["image/*"]},"x-ui-metadata":{"order":3,"advanced":false,"hidden":false,"is-mask":true,"mask-for":"image_urls","mask-mode":"alpha"}},"quality":{"type":"string","title":"Quality","description":"Rendering fidelity. \"High\" keeps typography crisp; use lower values only for previews.","default":"high","oneOf":[{"const":"low","title":"Low"},{"const":"medium","title":"Medium"},{"const":"high","title":"High"},{"const":"auto","title":"Auto"}],"x-ui-metadata":{"order":4,"advanced":true,"hidden":false}},"background":{"type":"string","title":"Background","description":"\"Opaque\" renders a filled background; \"Auto\" lets the model decide. Transparent backgrounds are not supported by this model.","default":"opaque","oneOf":[{"const":"opaque","title":"Opaque"},{"const":"auto","title":"Auto"}],"x-ui-metadata":{"order":5,"advanced":true,"hidden":false}},"moderation":{"type":"string","title":"Moderation","description":"Content filter strictness. \"Auto\" is the safe default.","default":"auto","oneOf":[{"const":"low","title":"Low"},{"const":"auto","title":"Auto"}],"x-ui-metadata":{"order":6,"advanced":true,"hidden":false}},"output_format":{"type":"string","title":"Output format","description":"\"PNG\" for crisp text, \"WEBP\" for smaller size, \"JPEG\" for broad compatibility.","default":"png","oneOf":[{"const":"png","title":"PNG"},{"const":"jpeg","title":"JPEG"},{"const":"webp","title":"WEBP"}],"x-ui-metadata":{"order":7,"advanced":true,"hidden":false}},"output_compression":{"type":"integer","title":"Output compression","description":"Compression level 0-100. 100 preserves text crispness; lower values reduce file size.","default":100,"minimum":0,"maximum":100,"x-ui-metadata":{"order":8,"advanced":true,"hidden":false}}},"required":["prompt"],"examples":[{"prompt":"A photorealistic, cinematic shot of a cozy independent bookstore in Mumbai at golden hour. Warm afternoon sunlight streams through a tall front window onto wooden shelves packed with books; book spines are clearly visible with titles in English and Hindi Devanagari script. In the foreground, a handwritten chalkboard A-frame easel reads, in clean legible chalk lettering: first line 'à¤®à¥à¤‚à¤¬à¤ˆ à¤ªà¥à¤¸à¥à¤¤à¤• à¤­à¤‚à¤¡à¤¾à¤°', second line 'Mumbai Book Store', and a smaller third line 'Open Daily  9 am â€“ 9 pm'. Shallow depth of field, shot on a full-frame camera at 35mm f/2.0, ultra-realistic detail, natural color grading, dust motes in sunlight, film grain.","size":"1536x1024","quality":"high"}]}},"metrics":{"total_requests":null,"latency":23.56594},"children":"$L2c"}]
23:{"metadata":[["$","title","0",{"children":"GPT Image 2 API | Segmind"}],["$","meta","1",{"name":"description","content":"Generate and edit photorealistic images with crisp multilingual typography, 2K resolution, and ~98% prompt adherence from a single API."}],["$","link","2",{"rel":"canonical","href":"https://www.segmind.com/models/gpt-image-2"}],["$","meta","3",{"property":"og:title","content":"GPT Image 2 API | Segmind"}],["$","meta","4",{"property":"og:description","content":"Generate and edit photorealistic images with crisp multilingual typography, 2K resolution, and ~98% prompt adherence from a single API."}],["$","meta","5",{"property":"og:image","content":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-3736.webp"}],["$","meta","6",{"name":"twitter:card","content":"summary_large_image"}],["$","meta","7",{"name":"twitter:title","content":"GPT Image 2 API | Segmind"}],["$","meta","8",{"name":"twitter:description","content":"Generate and edit photorealistic images with crisp multilingual typography, 2K resolution, and ~98% prompt adherence from a single API."}],["$","meta","9",{"name":"twitter:image","content":"https://www.segmind.com/og/home.jpg"}],["$","link","10",{"rel":"shortcut icon","href":"/favicon.ico"}],["$","link","11",{"rel":"icon","href":"/favicon.ico","sizes":"any"}],["$","link","12",{"rel":"icon","href":"/favicon-48x48.png","type":"image/png","sizes":"48x48"}],["$","link","13",{"rel":"icon","href":"/favicon-192x192.png","type":"image/png","sizes":"192x192"}],["$","link","14",{"rel":"icon","href":"/favicon.svg","type":"image/svg+xml"}],["$","link","15",{"rel":"apple-touch-icon","href":"/favicon-48x48.png"}],["$","$L2d","16",{}]],"error":null,"digest":"$undefined"}
27:"$23:metadata"
2c:["$","$L8",null,{"parallelRouterKey":"children","error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$La",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":"$undefined","forbidden":"$undefined","unauthorized":"$undefined"}]
2e:T6f6,{"@context":"https://schema.org","@type":"FAQPage","@id":"https://www.segmind.com/models/gpt-image-2#faq","mainEntity":[{"@type":"Question","name":"Does GPT Image 2 render text in Hindi, Japanese, and Chinese?","acceptedAnswer":{"@type":"Answer","text":"Yes. Multilingual typography is the model's flagship capability â€” Devanagari, CJK, Korean, and Bengali all render cleanly enough to ship."}},{"@type":"Question","name":"What is the difference between generation and edit mode?","acceptedAnswer":{"@type":"Answer","text":"Leaving the image parameter null generates from text alone. Passing an image URL switches the model into edit mode, where the prompt guides modifications to the input."}},{"@type":"Question","name":"What output sizes are supported?","acceptedAnswer":{"@type":"Answer","text":"1024x1024, 1536x1024, 1024x1536, and auto. All run up to 2K resolution with high quality."}},{"@type":"Question","name":"Does GPT Image 2 support transparent backgrounds?","acceptedAnswer":{"@type":"Answer","text":"No. background accepts opaque (the default) or auto; requesting transparent returns a 400. Use gpt-image-1.5 when you need an alpha channel for logos, stickers, or product cutouts."}},{"@type":"Question","name":"Is GPT Image 2 faster than gpt-image-1.5?","acceptedAnswer":{"@type":"Answer","text":"Yes â€” roughly 2Ã— faster thanks to a new single-pass architecture, with fewer artifacts on hands, faces, and material surfaces."}},{"@type":"Question","name":"Where does GPT Image 2 fall short?","acceptedAnswer":{"@type":"Answer","text":"Physical reasoning tasks (origami, angled reflections, Rubik's cubes) and highly dense repetitive detail (circuit diagrams, grains of sand) still challenge the model. Iterative edits beyond one or two passes tend to drift."}}]}1f:[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"SoftwareApplication\",\"@id\":\"https://www.segmind.com/models/gpt-image-2\",\"name\":\"GPT Image 2\",\"description\":\"Generate and edit photorealistic images with crisp multilingual typography, 2K resolution, and ~98% prompt adherence from a single API.\",\"url\":\"https://www.segmind.com/models/gpt-image-2\",\"applicationCategory\":\"AI Model\",\"applicationSubCategory\":\"textToImage\",\"operatingSystem\":\"Any\",\"image\":\"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-3736.webp\",\"offers\":{\"@type\":\"Offer\",\"availability\":\"https://schema.org/InStock\",\"priceCurrency\":\"USD\",\"price\":\"0.039338914469471985\"},\"provider\":{\"@type\":\"Organization\",\"@id\":\"https://www.segmind.com/#organization\",\"name\":\"Segmind\",\"url\":\"https://www.segmind.com\",\"logo\":\"https://www.segmind.com/favicon.svg\"}}"}}],["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"$2e"}}],["$","nav",null,{"aria-label":"Product pages","className":"px-4 pt-3 md:px-6","children":["$","ul",null,{"className":"flex flex-wrap gap-2","children":[["$","li","mcp",{"children":["$","$L1a",null,{"href":"/mcp","className":"inline-flex items-center gap-1 rounded-full border border-border px-3 py-1 text-sm text-muted-foreground transition-colors hover:text-foreground focus-visible:outline-2 focus-visible:outline-offset-2 focus-visible:outline-ring","children":["Segmind MCP Server: Generate Media from Any AI Agent",["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","stroke":"currentColor","strokeWidth":2,"strokeLinecap":"round","strokeLinejoin":"round","className":"lucide lucide-arrow-up-right size-3.5","aria-hidden":"true","children":["$L2f","$L30","$undefined"]}]]}]}]]}]}],"$L31"]
32:I[68382,["6693","static/chunks/fba4ba2d-460e04194f22506f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3009","static/chunks/6a7d804d-a5cc510ceffe206f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7505","static/chunks/c02d83f4-5637599209ef9447.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6541","static/chunks/fb4e7fc9-e31ad622f5fde25d.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9761","static/chunks/998e86a0-e112630ec4db17dc.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9509","static/chunks/9509-6125acd3006111c1.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4378","static/chunks/4378-3f82908971052aa9.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5289","static/chunks/5289-131715863bbab062.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4094","static/chunks/4094-eaa6158623fe27ff.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4982","static/chunks/4982-a4b2ee93ffd82ffd.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1810","static/chunks/1810-cf4ca0391c1ddf2a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6118","static/chunks/6118-7328b7d3f629a5ef.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","7979","static/chunks/7979-1799576a8e82a4f3.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","393","static/chunks/393-7cd2157fc3790475.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3985","static/chunks/3985-1f9bcb95de313d61.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","2610","static/chunks/2610-c41f461a4ed7cf24.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4844","static/chunks/4844-011fe3465334e7d0.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9305","static/chunks/9305-0f5d0f2a661afddc.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","713","static/chunks/713-dcb3ca4f0233704d.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","3504","static/chunks/3504-ae9bc2dc93d2ef47.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1742","static/chunks/1742-0ea812562873537a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1172","static/chunks/1172-89020d82938322e4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9919","static/chunks/9919-b585857c6e10fb85.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","9326","static/chunks/9326-103822841b140d80.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1885","static/chunks/1885-f206f5873f09b3e2.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","4077","static/chunks/4077-cba4e88bdb106467.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5126","static/chunks/5126-bdde68ce894e4935.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","5642","static/chunks/5642-c80cb7ee6e914a6a.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1924","static/chunks/1924-8c4ee98e9b0dbf83.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","6372","static/chunks/6372-3e2880fc7d769d1f.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1289","static/chunks/1289-4fdf72847f7ef615.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","8382","static/chunks/8382-004a14550e3769cc.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x","1951","static/chunks/app/models/%5Broute%5D/page-c5fd4df0a8a5cab4.js?dpl=dpl_CPYCfxesaX7RKfJUQZpH2JDib41x"],"default",1]
2f:["$","path","1tivn9",{"d":"M7 7h10v10"}]
30:["$","path","1vkiza",{"d":"M7 17 17 7"}]
33:Te48,# GPT Image 2: Photorealistic Text-to-Image and Edit Model

## What is GPT Image 2?

GPT Image 2 is OpenAI's next-generation image model, launched in April 2026 as the successor to gpt-image-1.5. It generates photorealistic images from text or edits existing images guided by a prompt, all through a single endpoint. The headline improvement is near-perfect in-image typography: over 95% accuracy across Latin, Japanese, Korean, Chinese, Hindi Devanagari, and Bengali scripts â€” the first image model practical for shipping UI labels, posters, and multilingual marketing assets without a manual redraw pass. A new single-pass architecture roughly doubles generation speed over the previous version, and built-in reasoning plans composition, counts items, and checks constraints before rendering.

## Key Features

- Text-to-image generation and guided image editing in one API
- In-image text rendering at 95%+ accuracy, including non-Latin scripts
- Output resolutions up to 2K across landscape (`1536x1024`), portrait (`1024x1536`), and square (`1024x1024`)
- Output formats: PNG (sharpest text), WebP (smaller files), JPEG (universal)
- Moderation controls: `auto` default, `low` for permitted use cases
- Native multi-constraint prompt adherence at ~98% accuracy

## Best Use Cases

GPT Image 2 is the right choice anytime legible text is part of the image: magazine covers with headlines, product packaging mockups, storefront and signboard scenes, infographics and charts, storyboards and comic panels, multilingual ad creatives, UI screen mockups, and posters. In testing, it rendered a handwritten chalkboard easel combining English ("Mumbai Book Store", "Open Daily 9 am â€“ 9 pm") and Hindi Devanagari ("à¤®à¥à¤‚à¤¬à¤ˆ à¤ªà¥à¤¸à¥à¤¤à¤• à¤­à¤‚à¤¡à¤¾à¤°") cleanly on the first try. Edit mode (passing an `image` input) is ideal for relighting, background swaps, text changes on existing visuals, and brand-consistent variations.

## Prompt Tips and Output Quality

Keep `quality=high` whenever typography matters â€” `medium` and `low` degrade fine lettering. Lead the prompt with subject, then typography in quotes, then style and lighting cues. For magazine-style layouts pick `1024x1536`; for marketing banners and scenes, `1536x1024`. Keep `output_format=png` and `output_compression=100` when text crispness is non-negotiable.

## FAQs

**Does GPT Image 2 render text in Hindi, Japanese, and Chinese?**
Yes. Multilingual typography is the model's flagship capability â€” Devanagari, CJK, Korean, and Bengali all render cleanly enough to ship.

**What is the difference between generation and edit mode?**
Leaving the `image` parameter null generates from text alone. Passing an image URL switches the model into edit mode, where the prompt guides modifications to the input.

**What output sizes are supported?**
`1024x1024`, `1536x1024`, `1024x1536`, and `auto`. All run up to 2K resolution with high quality.

**Does GPT Image 2 support transparent backgrounds?**
No. `background` accepts `opaque` (the default) or `auto`; requesting `transparent` returns a 400. Use `gpt-image-1.5` when you need an alpha channel for logos, stickers, or product cutouts.

**Is GPT Image 2 faster than gpt-image-1.5?**
Yes â€” roughly 2Ã— faster thanks to a new single-pass architecture, with fewer artifacts on hands, faces, and material surfaces.

**Where does GPT Image 2 fall short?**
Physical reasoning tasks (origami, angled reflections, Rubik's cubes) and highly dense repetitive detail (circuit diagrams, grains of sand) still challenge the model. Iterative edits beyond one or two passes tend to drift.
34:T951,# How to Use GPT Image 2

GPT Image 2 handles both text-to-image generation and guided editing through a single endpoint. The way you configure its parameters decides whether output lands at magazine-ready quality or drifts into preview territory â€” especially when in-image typography is part of the brief.

## Structuring the Prompt

Lead with the subject and scene, then the exact text you want rendered (in quotes), then style and lighting cues. Specify fonts, scripts, and hierarchy when shipping multilingual creatives. Example: *"A chalkboard easel reads 'Open Daily 9 am â€“ 9 pm' in clean white chalk, shallow depth of field, golden-hour lighting."* The model handles Latin, Japanese, Korean, Chinese, Hindi Devanagari, and Bengali without script switching tricks.

## Recommended Settings

Based on working test runs, these defaults produce production-ready output:

- **`quality=high`** â€” the single most important setting. Drop it only for quick previews; `medium` and `low` visibly degrade lettering.
- **`size`** â€” `1536x1024` for marketing banners and scene compositions; `1024x1536` for magazine covers, posters, and portrait social; `1024x1024` for in-feed social.
- **`output_format=png`** with `output_compression=100` whenever text crispness matters. Switch to `webp` for web delivery where file size is a priority.
- **`moderation=auto`** is the safe default; switch to `low` only for permitted use cases.
- **`background=opaque`** (the default) or `auto` â€” this model does not support transparent output. For cutouts that need an alpha channel, use `gpt-image-1.5`.

## Text-to-Image vs Edit Mode

Leave the `image` parameter null for pure text-to-image generation. To edit an existing image â€” relighting, background swap, on-image text change, brand variations â€” pass the image URL and the model switches into edit mode automatically. The same prompt guidelines apply: be specific about what you want changed and what must be preserved.

## Best Practices

Batch multi-constraint prompts (object counts, colors, text strings, positions) in a single pass â€” GPT Image 2 resolves around 98% of these without rerolls. Avoid iterative editing beyond one or two passes; quality plateaus and drift creeps in. For dense technical diagrams or physically constrained geometry (origami, reflections), validate outputs before shipping.
35:T8ae,## How Pricing Works

GPT Image 2 is billed per token: $8.00 per million input tokens (your prompt and any reference image) and $30.00 per million output tokens (the generated image). Output cost dominates in almost every real workload, so the `quality` parameter and the output `size` are the two levers that actually move the bill. All other parameters (`output_format`, `output_compression`, `background`, `moderation`) are free.

### Pricing Examples

Costs below are per call, measured from live API header deductions on April 22, 2026.

| Size | Low Quality | Medium Quality | High Quality |
|---|---|---|---|
| 1024 x 768 | $0.01 | $0.04 | $0.15 |
| 1024 x 1024 | $0.01 | $0.06 | $0.22 |
| 1024 x 1536 | $0.01 | $0.05 | $0.17 |
| 1920 x 1080 | $0.01 | $0.04 | $0.16 |
| 2560 x 1440 | $0.01 | $0.06 | $0.23 |
| 3840 x 2160 | $0.02 | $0.11 | $0.41 |

### Cost Optimization Tips

- Use `quality=low` or `quality=medium` for ideation passes. At `low` quality every supported size is $0.01â€“$0.02 per image, so you can burn through 15â€“20 exploratory prompts for the cost of one `high` 4K render.
- `1024x768`, `1024x1024`, and `1920x1080` are the cheapest production sizes at `high` quality ($0.15â€“$0.22). Reach for `2560x1440` or `3840x2160` only when delivery resolution actually demands it â€” 4K `high` is nearly 3x the cost of 1080p `high`.
- `output_format` and `output_compression` do not change the price. Pick `jpeg` at compression 90 for CDN delivery, `png` at 100 when text crispness is non-negotiable.
- Edit mode (passing an `image` input) adds input tokens for the reference. Keep reference images under 1024px on the longest edge unless you need the detail.

### Example Scenario

A marketing agency shipping 20 localized Instagram Story ad variants per week at `1024x1536` and `quality=high` pays roughly $0.17 per image, or $3.40 per week. A production house generating 500 concert poster drafts per month at `quality=medium` (1024x1024) before selecting finalists to re-render at `high` lands near $40 per month, all-in. Teams delivering 4K hero assets at `3840x2160` `quality=high` should budget $0.41 per image â€” roughly 2.4x the cost of the same shot at 1024x1536.
31:["$","$L32",null,{"route":"gpt-image-2","modelDetails":{"title":"GPT Image 2","thumbnail_image":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-3736.webp","type":"textToImage","model_type":"SO","is_depreciated":false,"meta_description":"Generate and edit photorealistic images with crisp multilingual typography, 2K resolution, and ~98% prompt adherence from a single API.","seo_description":"$33","guide":"$34","pricing_guide":"$35","artifacts":"$1e:props:modelDetails:artifacts","default_image_output":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-example-output.png","default_response":null,"parameters":"$1e:props:modelDetails:parameters","endpoints":"$1e:props:modelDetails:endpoints","api_notes":"$1e:props:modelDetails:api_notes","model_outputs":null,"llm_capabilities":null,"license":"","github_link":null,"inference_provider":"OPENAI","created_ts":"2026-04-22 05:04:46.746223+00:00","slug":"gpt-image-2","name":"GPT Image 2","description":"Generate photorealistic images with legible multilingual text and 2K output.","category":"textToImage","modality":"image","input":"text","output":"image","provider":"OPENAI","thumbnail_url":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2-3736.webp","is_new":false,"coming_soon":false,"popularity_score":91,"category_percentile":99,"average_cost":0.039338914469471985,"latency_seconds":23.56594,"created_at":"2026-04-22 05:04:46.746223+00:00","parameters_schema":"$1e:props:modelDetails:parameters_schema"},"relatedTemplates":[],"popularModels":[{"slug":"gpt-image-2.5-sunburst","title":"GPT Image 2.5 Sunburst","description":"Precisely edit and generate images with legible in-image text.","type":"textToImage","partner":"OPENAI","thumbnail_image":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2.5-sunburst-8190.webp","default_image_output":"","is_new":false,"coming_soon":false,"is_visible":true,"segmind":true,"is_depreciated":false,"parameters":null,"latency":40.45285,"average_cost":0.09146240173350827,"created_at":"2026-09-09 03:35:22.170019+00:00","popularity_score":78,"popularity_scale":"score","category_percentile":95},{"slug":"gpt-image-2.5-flare","title":"GPT Image 2.5 Flare","description":"Fast text-to-image and editing with legible in-image text.","type":"textToImage","partner":"OPENAI","thumbnail_image":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-2.5-flare-4502.webp","default_image_output":"","is_new":false,"coming_soon":false,"is_visible":true,"segmind":true,"is_depreciated":false,"parameters":null,"latency":17.85295,"average_cost":0.03886577973894409,"created_at":"2026-09-09 03:35:21.281821+00:00","popularity_score":71,"popularity_scale":"score","category_percentile":93},{"slug":"gpt-image-1.5","title":"GPT Image 1.5","description":"Stunning photorealistic images with exceptional instruction-following.","type":"textToImage","partner":"OPENAI","thumbnail_image":"https://segmind-resources.s3.amazonaws.com/output/gpt-image-1.5-6628.png","default_image_output":"","is_new":false,"coming_soon":false,"is_visible":true,"segmind":true,"is_depreciated":false,"parameters":null,"latency":31.62128,"average_cost":0.17231282649253732,"created_at":"2025-12-17 19:02:37.492211+00:00","popularity_score":58,"popularity_scale":"score","category_percentile":79},{"slug":"gpt-image-1","title":"GPT Image 1","description":"Create high-quality AI-generated images from text prompts using OpenAI's GPT Image 1 model. Ideal for product design, content creation, and rapid visual prototyping at scale.","type":"textToImage","partner":"OPENAI","thumbnail_image":"https://segmind-resources.s3.amazonaws.com/input/34337ae8-1317-4afb-ace6-ae23d4bdb12e-ChatGPT_Image_Apr_24_2025_01_37_13_AM_1.png","default_image_output":"","is_new":false,"coming_soon":false,"is_visible":true,"segmind":true,"is_depreciated":false,"parameters":null,"latency":37.20149,"average_cost":0.20091909838350053,"created_at":"2025-04-23 19:47:45.737469+00:00","popularity_score":63,"popularity_scale":"score","category_percentile":86}]}]
