Buckets:
| <meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Planificadores","local":"schedulers","sections":[{"title":"Elegir un programador","local":"choosing-a-scheduler","sections":[],"depth":2},{"title":"Carga de programadores","local":"loading-schedulers","sections":[],"depth":2},{"title":"Programas de pasos de tiempo","local":"timestep-schedules","sections":[],"depth":2},{"title":"Espaciado de pasos de tiempo","local":"timestep-spacing","sections":[],"depth":2},{"title":"Sigmas","local":"sigmas","sections":[{"title":"Sigmas de Karras","local":"karras-sigmas","sections":[],"depth":3}],"depth":2},{"title":"Próximos pasos","local":"next-steps","sections":[],"depth":2}],"depth":1}"/> | |
| <link href="/docs/diffusers/main/en/_app/immutable/entry/start.0S0_iEoG.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/I9eooN77.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/Bupkjr9f.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/entry/app.CPdOxCbw.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/CyvZ3FS5.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/l0QnnOED.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/DnVIJfC5.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/nodes/0.DteTvmQZ.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/CoSrdQax.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/nodes/406.C280A7QI.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/DSPDl1gz.js" rel="modulepreload"> | |
| <!--alcvz4--><meta name="hf:doc:metadata" content="{"title":"Schedulers","local":"schedulers","sections":[{"title":"Choosing a scheduler","local":"choosing-a-scheduler","sections":[],"depth":2},{"title":"Loading schedulers","local":"loading-schedulers","sections":[],"depth":2},{"title":"Timestep schedules","local":"timestep-schedules","sections":[],"depth":2},{"title":"Timestep spacing","local":"timestep-spacing","sections":[],"depth":2},{"title":"Sigmas","local":"sigmas","sections":[{"title":"Karras sigmas","local":"karras-sigmas","sections":[],"depth":3}],"depth":2},{"title":"Next steps","local":"next-steps","sections":[],"depth":2}],"depth":1}"/><!----> | |
| <link href="/docs/diffusers/main/en/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: inline-end; margin-inline-start: 10px; display: inline-flex; position: relative; z-index: 10;"><div dir="ltr" class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----></span> <span dir="auto">Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg><!----></button></div> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="schedulers" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#schedulers"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Planificadores<!----></span></h1><!--]--><!----> <p>Un programador indica al bucle de desruidoado cuánto ruido eliminar en cada paso. Los distintos programadores equilibran la velocidad con la calidad.</p> <p>Esta guía muestra cómo cargar un programador y personalizar su programación de pasos de tiempo, su espaciado y sus sigmas.</p> <!--[1--><h2 class="relative group"><a id="choosing-a-scheduler" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#choosing-a-scheduler"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Elegir un programador<!----></span></h2><!--]--><!----> <p>Comience desde el punto de control predeterminado. Cambie solo si necesita un equilibrio diferente entre velocidad y calidad.</p> <ul><li>DPM++ 2M SDE Karras es una opción versátil y eficaz para muchos puntos de control de difusión latente.</li> <li><a href="/docs/diffusers/main/es/api/schedulers/tcd#diffusers.TCDScheduler">TCDScheduler</a> funciona bien con modelos destilados.</li> <li>Usa <a href="/docs/diffusers/main/es/api/schedulers/flow_match_euler_discrete#diffusers.FlowMatchEulerDiscreteScheduler">FlowMatchEulerDiscreteScheduler</a> o <a href="/docs/diffusers/main/es/api/schedulers/flow_match_heun_discrete#diffusers.FlowMatchHeunDiscreteScheduler">FlowMatchHeunDiscreteScheduler</a> para modelos FlowMatch (Qwen-Image, Flux y similares).</li> <li><a href="/docs/diffusers/main/es/api/schedulers/euler#diffusers.EulerDiscreteScheduler">EulerDiscreteScheduler</a> o <a href="/docs/diffusers/main/es/api/schedulers/euler_ancestral#diffusers.EulerAncestralDiscreteScheduler">EulerAncestralDiscreteScheduler</a> suelen funcionar bien para imágenes de estilo anime.</li> <li><a href="/docs/diffusers/main/es/api/schedulers/lcm#diffusers.LCMScheduler">LCMScheduler</a> con un UNet LCM o un LoRA para la generación en pocos pasos cuando el punto de control lo admite.</li></ul> <!--[1--><h2 class="relative group"><a id="loading-schedulers" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#loading-schedulers"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Carga de programadores<!----></span></h2><!--]--><!----> <blockquote class="tip"><p>Los modelos de flow-matching como Qwen-Image y Flux incluyen <a href="/docs/diffusers/main/es/api/schedulers/flow_match_euler_discrete#diffusers.FlowMatchEulerDiscreteScheduler">FlowMatchEulerDiscreteScheduler</a> como valor predeterminado. Mantén ese scheduler a menos que estés experimentando de forma intencional. Sustitúyelo mediante <a href="/docs/diffusers/main/es/api/configuration#diffusers.ConfigMixin.from_config">from_config()</a> solo cuando el reemplazo sea compatible con el checkpoint.</p></blockquote> <p>Los Schedulers solo contienen configuración y no incluyen tensores de pesos. Acceda al atributo <code>.scheduler</code> de un pipeline para inspeccionar la configuración cargada.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch | |
| <span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline | |
| pipeline = DiffusionPipeline.from_pretrained( | |
| <span class="hljs-string">"stabilityai/stable-diffusion-xl-base-1.0"</span>, dtype=torch.float16, device_map=<span class="hljs-string">"cuda"</span> <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| ) | |
| pipeline.scheduler<!----></pre></div><!----> <p>Para cambiar los programadores en una pipeline cargada, use <a href="/docs/diffusers/main/es/api/configuration#diffusers.ConfigMixin.from_config">from_config()</a> con la configuración del programador existente para que <code>num_train_timesteps</code> y los campos relacionados permanezcan alineados. Para los puntos de control de FlowMatch (Qwen-Image, Flux y similares), mantenga <a href="/docs/diffusers/main/es/api/schedulers/flow_match_euler_discrete#diffusers.FlowMatchEulerDiscreteScheduler">FlowMatchEulerDiscreteScheduler</a> a menos que esté experimentando intencionalmente con un reemplazo compatible.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DPMSolverMultistepScheduler | |
| pipeline.scheduler = DPMSolverMultistepScheduler.from_config(pipeline.scheduler.config)<!----></pre></div><!----> <p>También puedes cargar una configuración de programador desde el Hub con <a href="/docs/diffusers/main/es/api/schedulers/overview#diffusers.SchedulerMixin.from_pretrained">from_pretrained()</a> y pasarla a <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pretrained">from_pretrained()</a> a través de <code>scheduler</code>.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, DPMSolverMultistepScheduler | |
| dpm = DPMSolverMultistepScheduler.from_pretrained( | |
| <span class="hljs-string">"stabilityai/stable-diffusion-xl-base-1.0"</span>, subfolder=<span class="hljs-string">"scheduler"</span> | |
| ) | |
| pipeline = DiffusionPipeline.from_pretrained( | |
| <span class="hljs-string">"stabilityai/stable-diffusion-xl-base-1.0"</span>, | |
| scheduler=dpm, | |
| dtype=torch.float16, | |
| device_map=<span class="hljs-string">"cuda"</span>, <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| ) | |
| pipeline.scheduler<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="timestep-schedules" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#timestep-schedules"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Programas de pasos de tiempo<!----></span></h2><!--]--><!----> <p>El timestep o la programación de ruido determina cómo se distribuye el ruido a lo largo del proceso de desruido. La programación puede ser lineal o más concentrada hacia el inicio o el final. Es una secuencia precalculada de niveles de ruido generada a partir de la configuración predeterminada del scheduler, pero se puede personalizar para utilizar otras programaciones.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-text "><!---->linear (even steps) AYS (denser where it matters) | |
| noise noise | |
| ^ ^ | |
| | * | * | |
| | * | * | |
| | * | ** | |
| | * | *** | |
| | * | ** | |
| | * | * | |
| | * | * | |
| +-----------------> step +-----------------> step<!----></pre></div><!----> <blockquote class="tip"><p>Los <code>timesteps</code> personalizados solo funcionan si el <code>set_timesteps</code> de ese scheduler acepta el argumento (los pipelines comprueban la firma mediante <code>retrieve_timesteps</code> y lanzan <code>ValueError</code> en caso contrario). Consulta la página de API del scheduler o la firma de <code>set_timesteps</code> antes de pasarlos.</p></blockquote> <p>El ejemplo siguiente utiliza el programa <a href="https://research.nvidia.com/labs/toronto-ai/AlignYourSteps/" rel="nofollow">Align Your Steps (AYS)</a>, que puede generar una imagen de alta calidad en 10 pasos, acelerando significativamente la generación y reduciendo el tiempo de cómputo.</p> <p>Importa el schedule y pásalo al argumento <code>timesteps</code> en el pipeline.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch | |
| <span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, DPMSolverMultistepScheduler | |
| <span class="hljs-keyword">from</span> diffusers.schedulers <span class="hljs-keyword">import</span> AysSchedules | |
| sampling_schedule = AysSchedules[<span class="hljs-string">"StableDiffusionXLTimesteps"</span>] | |
| <span class="hljs-built_in">print</span>(sampling_schedule) | |
| <span class="hljs-comment"># [999, 845, 730, 587, 443, 310, 193, 116, 53, 13]</span> | |
| pipeline = DiffusionPipeline.from_pretrained( | |
| <span class="hljs-string">"SG161222/RealVisXL_V4.0"</span>, | |
| dtype=torch.float16, | |
| device_map=<span class="hljs-string">"cuda"</span> <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| ) | |
| pipeline.scheduler = DPMSolverMultistepScheduler.from_config( | |
| pipeline.scheduler.config, algorithm_type=<span class="hljs-string">"sde-dpmsolver++"</span> | |
| ) | |
| prompt = <span class="hljs-string">"A cinematic shot of a cute little rabbit wearing a jacket and doing a thumbs up"</span> | |
| image = pipeline( | |
| prompt=prompt, | |
| negative_prompt=<span class="hljs-string">""</span>, | |
| timesteps=sampling_schedule, | |
| ).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div class="flex gap-4"><div><img class="rounded-xl" src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ays.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">Programa de pasos de tiempo AYS 10 pasos</figcaption></div> <div><img class="rounded-xl" src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/10.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">Programa de pasos de tiempo espaciados linealmente 10 pasos</figcaption></div> <div><img class="rounded-xl" src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/25.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">Programa de pasos de tiempo espaciados linealmente 25 pasos</figcaption></div></div> <p>Para el reescalado de v-prediction/zero-SNR en puntos de control antiguos, consulte <a href="./legacy_checkpoints#rescaling-schedules">Puntos de control heredados</a>.</p> <!--[1--><h2 class="relative group"><a id="timestep-spacing" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#timestep-spacing"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Espaciado de pasos de tiempo<!----></span></h2><!--]--><!----> <p>La separación de pasos de tiempo se refiere a los pasos específicos <em>t</em> desde los que muestrear en la programación. Diffusers proporciona tres tipos de separación, como se muestra a continuación.</p> <table><thead><tr><th>estrategia de espaciado</th><th>cálculo del espaciado</th><th>pasos de tiempo de ejemplo</th></tr></thead><tbody><tr><td><code>leading</code></td><td>pasos equidistantes</td><td><code>[900, 800, 700, ..., 100, 0]</code></td></tr><tr><td><code>linspace</code></td><td>incluye los pasos inicial y final y divide uniformemente los pasos intermedios restantes</td><td><code>[1000, 888.89, 777.78, ..., 111.11, 0]</code></td></tr><tr><td><code>trailing</code></td><td>incluye el último paso y divide uniformemente los pasos intermedios restantes comenzando desde el final</td><td><code>[999, 899, 799, 699, 599, 499, 399, 299, 199, 99]</code></td></tr></tbody></table> <p>Pasa la estrategia de espaciado al argumento <code>timestep_spacing</code> en el programador.</p> <blockquote class="tip"><p>La estrategia <code>trailing</code> suele producir imágenes de mayor calidad y con más detalles en menos pasos, pero la diferencia de calidad no es tan evidente para valores de pasos más estándar.</p></blockquote> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch | |
| <span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, DPMSolverMultistepScheduler | |
| pipeline = DiffusionPipeline.from_pretrained( | |
| <span class="hljs-string">"SG161222/RealVisXL_V4.0"</span>, | |
| dtype=torch.float16, | |
| device_map=<span class="hljs-string">"cuda"</span> <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| ) | |
| pipeline.scheduler = DPMSolverMultistepScheduler.from_config( | |
| pipeline.scheduler.config, timestep_spacing=<span class="hljs-string">"trailing"</span> | |
| ) | |
| prompt = <span class="hljs-string">"A cinematic shot of a cute little black cat sitting on a pumpkin at night"</span> | |
| image = pipeline( | |
| prompt=prompt, | |
| negative_prompt=<span class="hljs-string">""</span>, | |
| num_inference_steps=<span class="hljs-number">5</span>, | |
| ).images[<span class="hljs-number">0</span>] | |
| image<!----></pre></div><!----> <div class="flex gap-4"><div><img class="rounded-xl" src="https://huggingface.co/datasets/stevhliu/testing-images/resolve/main/trailing_spacing.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">espaciado final tras 5 pasos</figcaption></div> <div><img class="rounded-xl" src="https://huggingface.co/datasets/stevhliu/testing-images/resolve/main/leading_spacing.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">espaciado inicial tras 5 pasos</figcaption></div></div> <!--[1--><h2 class="relative group"><a id="sigmas" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#sigmas"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Sigmas<!----></span></h2><!--]--><!----> <p>Sigmas es una medida de lo ruidoso que es una muestra en un paso determinado, según lo definido por la programación. Al usar <code>sigmas</code> personalizados, los <code>timesteps</code> se calculan a partir de estos valores en lugar de la configuración predeterminada del programador.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-text "><!---->step: 0 1 2 3 4 | |
| sigma: σ0 > σ1 > σ2 > σ3 > σ4 ≈ 0 | |
| high noise ---> clean sample<!----></pre></div><!----> <blockquote class="tip"><p>Los <code>sigmas</code> personalizados solo funcionan si el <code>set_timesteps</code> de ese programador acepta el argumento (los pipelines comprueban la firma mediante <code>retrieve_timesteps</code> y lanzan <code>ValueError</code> en caso contrario). Consulta la página de API del programador o la firma de <code>set_timesteps</code> antes de pasarlos.</p></blockquote> <p>Pasa los sigmas personalizados al argumento <code>sigmas</code> en el pipeline. El ejemplo de abajo usa los <a href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/schedulers/scheduling_utils.py" rel="nofollow">sigmas</a> de la programación AYS de 10 pasos.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch | |
| <span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, EulerDiscreteScheduler | |
| <span class="hljs-keyword">from</span> diffusers.schedulers <span class="hljs-keyword">import</span> AysSchedules | |
| pipeline = DiffusionPipeline.from_pretrained( | |
| <span class="hljs-string">"SG161222/RealVisXL_V4.0"</span>, | |
| dtype=torch.float16, | |
| device_map=<span class="hljs-string">"cuda"</span>, <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| ) | |
| pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config) | |
| sigmas = AysSchedules[<span class="hljs-string">"StableDiffusionXLSigmas"</span>] | |
| prompt = <span class="hljs-string">"A cinematic shot of a cute little rabbit wearing a jacket and doing a thumbs up"</span> | |
| image = pipeline( | |
| prompt=prompt, | |
| negative_prompt=<span class="hljs-string">""</span>, | |
| sigmas=sigmas, | |
| ).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <!--[2--><h3 class="relative group"><a id="karras-sigmas" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#karras-sigmas"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Sigmas de Karras<!----></span></h3><!--]--><!----> <p><a href="https://huggingface.co/papers/2206.00364" rel="nofollow">Karras sigmas</a> re-muestrea el plan de ruido para un muestreo más eficiente, agrupando los sigmas de forma más densa en la parte media de la secuencia, donde la reconstrucción de la estructura es crítica, mientras que utiliza menos sigmas al principio y al final, donde los cambios de ruido tienen menos impacto. Esto puede aumentar el nivel de detalle en una imagen generada.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-text "><!---->default σ: * * * * * * * * even-ish | |
| Karras σ: * * * * * * * denser mid, sparser ends | |
| |---structure---|<!----></pre></div><!----> <p>Establezca <code>use_karras_sigmas=True</code> en el programador para habilitarlo.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch | |
| <span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, DPMSolverMultistepScheduler | |
| pipeline = DiffusionPipeline.from_pretrained( | |
| <span class="hljs-string">"SG161222/RealVisXL_V4.0"</span>, | |
| dtype=torch.float16, | |
| device_map=<span class="hljs-string">"cuda"</span> <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| ) | |
| pipeline.scheduler = DPMSolverMultistepScheduler.from_config( | |
| pipeline.scheduler.config, | |
| algorithm_type=<span class="hljs-string">"sde-dpmsolver++"</span>, | |
| use_karras_sigmas=<span class="hljs-literal">True</span>, | |
| ) | |
| prompt = <span class="hljs-string">"A cinematic shot of a cute little rabbit wearing a jacket and doing a thumbs up"</span> | |
| image = pipeline( | |
| prompt=prompt, | |
| negative_prompt=<span class="hljs-string">""</span>, | |
| num_inference_steps=<span class="hljs-number">20</span>, | |
| ).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div class="flex gap-4"><div><img class="rounded-xl" src="https://huggingface.co/datasets/stevhliu/testing-images/resolve/main/karras_sigmas_true.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">Karras sigmas habilitado</figcaption></div> <div><img class="rounded-xl" src="https://huggingface.co/datasets/stevhliu/testing-images/resolve/main/karras_sigmas_false.png"/> <figcaption class="mt-2 text-center text-sm text-gray-500">Karras sigmas desactivado</figcaption></div></div> <p>Consulte el <a href="../api/schedulers/overview">resumen</a> de la API del programador para obtener una lista de programadores que admiten sigmas de Karras. Solo debe usarse con modelos entrenados con sigmas de Karras.</p> <!--[1--><h2 class="relative group"><a id="next-steps" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#next-steps"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Próximos pasos<!----></span></h2><!--]--><!----> <ul><li>Lea el artículo <a href="https://huggingface.co/papers/2305.08891" rel="nofollow">Common Diffusion Noise Schedules and Sample Steps are Flawed</a> para obtener más detalles sobre el reescalado del ruido para aplicar un SNR cero.</li></ul> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/diffusers/blob/main/docs/source/en/using-diffusers/schedulers.md" target="_blank"><svg class="mr-1" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg><!----> <span><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]--> | |
| <script> | |
| { | |
| __sveltekit_1qsdbwf = { | |
| base: "/docs/diffusers/main/en", | |
| assets: "/docs/diffusers/main/en" | |
| }; | |
| const element = document.currentScript.parentElement; | |
| Promise.all([ | |
| import("/docs/diffusers/main/en/_app/immutable/entry/start.0S0_iEoG.js"), | |
| import("/docs/diffusers/main/en/_app/immutable/entry/app.CPdOxCbw.js") | |
| ]).then(([kit, app]) => { | |
| kit.start(app, element, { | |
| node_ids: [0, 406], | |
| data: [null,null], | |
| form: null, | |
| error: null | |
| }); | |
| }); | |
| } | |
| </script> | |
Xet Storage Details
- Size:
- 44.5 kB
- Xet hash:
- c6c76936bacc2b516ddad0094bbe0835454631342f9fa6476a444a1d150fb090
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.