Buckets:

hf-doc-build
/

doc

Files
xet
hf-doc-build/doc / diffusers /main /es /api /pipelines /flux3.html
mishig
about 3 hours ago
download
raw
97.2 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;FLUX 3 Image Klein&quot;,&quot;local&quot;:&quot;flux-3-image-klein&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Ampliación de prompts&quot;,&quot;local&quot;:&quot;prompt-upsampling&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Imagen a imagen&quot;,&quot;local&quot;:&quot;image-to-image&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Control de diseño&quot;,&quot;local&quot;:&quot;layout-control&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Escala de guía&quot;,&quot;local&quot;:&quot;guidance-scale&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Modelo destilado por pasos&quot;,&quot;local&quot;:&quot;step-distilled-model&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Generación reproducible&quot;,&quot;local&quot;:&quot;reproducible-generation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flujos de trabajo&quot;,&quot;local&quot;:&quot;workflows&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Callbacks por paso&quot;,&quot;local&quot;:&quot;per-step-callbacks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Memoria&quot;,&quot;local&quot;:&quot;memory&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Carga de un solo archivo&quot;,&quot;local&quot;:&quot;single-file-loading&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3ModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.Flux3ModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3AutoBlocks&quot;,&quot;local&quot;:&quot;diffusers.Flux3AutoBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3DistilledAutoBlocks&quot;,&quot;local&quot;:&quot;diffusers.Flux3DistilledAutoBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3DistilledModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.Flux3DistilledModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
		<link href="/docs/diffusers/main/en/_app/immutable/entry/start.U6TUchwJ.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/yxSHiGDF.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/CPfOwUO_.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/entry/app.3qKhPoX-.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/BYe-f74j.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/chunks/D32hXVzh.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/0.DqpuBF3n.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Dyp9xKq8.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/8.AKHyGgGH.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Dk44y1gU.js" rel="modulepreload">
		<!--hchutw--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;FLUX 3 Image Klein&quot;,&quot;local&quot;:&quot;flux-3-image-klein&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Prompt upsampling&quot;,&quot;local&quot;:&quot;prompt-upsampling&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Image-to-image&quot;,&quot;local&quot;:&quot;image-to-image&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Layout control&quot;,&quot;local&quot;:&quot;layout-control&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Guidance scale&quot;,&quot;local&quot;:&quot;guidance-scale&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Step-distilled model&quot;,&quot;local&quot;:&quot;step-distilled-model&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Reproducible generation&quot;,&quot;local&quot;:&quot;reproducible-generation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Workflows&quot;,&quot;local&quot;:&quot;workflows&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Per-step callbacks&quot;,&quot;local&quot;:&quot;per-step-callbacks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Memory&quot;,&quot;local&quot;:&quot;memory&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Single-file loading&quot;,&quot;local&quot;:&quot;single-file-loading&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3ModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.Flux3ModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3AutoBlocks&quot;,&quot;local&quot;:&quot;diffusers.Flux3AutoBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3DistilledAutoBlocks&quot;,&quot;local&quot;:&quot;diffusers.Flux3DistilledAutoBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Flux3DistilledModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.Flux3DistilledModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/><!---->
		<link href="/docs/diffusers/main/en/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <!--[0--><h1 class="relative group"><a id="flux-3-image-klein" 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="#flux-3-image-klein"><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><!---->FLUX 3 Image Klein<!----></span></h1><!--]--><!----> <p>FLUX 3 Image Klein es un modelo de generación de imágenes de 5.6B de Black Forest Labs: un único transformador (<a href="/docs/diffusers/main/es/api/models/flux3_transformer#diffusers.Flux3Transformer2DModel">Flux3Transformer2DModel</a>) genera imágenes a partir de un prompt de texto, solo o junto con una o más imágenes de referencia. La condicionación de texto proviene de Qwen3-VL, y las imágenes utilizan el autoencoder empaquetado de FLUX.2 (<code>AutoencoderKLFlux2</code>).</p> <p>FLUX 3 Image Klein se distribuye como un <a href="../../modular_diffusers/overview">pipeline modular</a>. Comienza con una generación de imagen a partir de texto:</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-python "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> ModularPipeline

pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)

image = pipe(
    prompt=<span class="hljs-string">&quot;A red fox in a snow-covered forest clearing at golden hour&quot;</span>,
    generator=torch.Generator().manual_seed(<span class="hljs-number">0</span>),
    output=<span class="hljs-string">&quot;images&quot;</span>,
)[<span class="hljs-number">0</span>]
image.save(<span class="hljs-string">&quot;fox.png&quot;</span>)<!----></pre></div><!----> <p><code>prompt</code> es la única entrada obligatoria: el pipeline aplica la configuración de muestreo recomendada (50 pasos, escala de guía 3.5, desplazamiento del programador 6.93) y genera imágenes de 1024 x 1024 píxeles de forma predeterminada. Acepta un solo prompt por llamada; use <code>num_images_per_prompt</code> para generar varias imágenes a partir del mismo prompt.</p> <p>Como puede ver en el ejemplo anterior, la carga y la ejecución difieren un poco de un <code>DiffusionPipeline</code>. La carga es perezosa: <a href="/docs/diffusers/main/es/api/modular_diffusers/pipeline#diffusers.ModularPipeline.from_pretrained">from_pretrained()</a> solo construye el pipeline, y llama a <a href="/docs/diffusers/main/es/api/modular_diffusers/pipeline#diffusers.ModularPipeline.load_components">load_components()</a> para cargar los pesos. Una llamada devuelve el estado del pipeline, y solicita lo que necesita con <code>output</code>: <code>output="images"</code> aquí, o <code>output=["images", "prompt"]</code> para obtener también la descripción que escribió el amplificador de prompts.</p> <blockquote class="tip"><p>Si estás trabajando con <code>ModularPipeline</code> por primera vez, lee el <a href="../../modular_diffusers/quickstart">quickstart</a> y <a href="../../modular_diffusers/coming_from_diffusion_pipeline">Coming from DiffusionPipeline</a> si conoces las pipelines estándar. Además de la carga y las salidas, hay algunas otras diferencias: <a href="#guidance-scale">guidance</a>, <a href="#per-step-callbacks">callbacks</a> y <a href="#memory">offloading</a>.</p></blockquote> <!--[1--><h2 class="relative group"><a id="prompt-upsampling" 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="#prompt-upsampling"><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><!---->Ampliación de prompts<!----></span></h2><!--]--><!----> <p>Prompting matters: FLUX 3 Image Klein takes full scene captions that describe the subject, composition, lighting, and atmosphere, not short prompts; a request sent as typed gives generic output. That is why the pipeline runs the prompt upsampler first on every request, text-to-image and editing alike. It is a separate 9B model from Black Forest Labs (<a href="https://huggingface.co/black-forest-labs/flux-3-image-klein-pup" rel="nofollow">black-forest-labs/flux-3-image-klein-pup</a>) that expands <code>prompt</code> into such a caption before the text encoder, and a short prompt is all it takes.</p> <p>El amplificador de prompts convierte una solicitud breve, junto con las imágenes de referencia en caso de edición, en la descripción completa de la escena que recibe el modelo: medio y estilo, escena, iluminación y cámara, cada sujeto con sus colores, materiales y posición en el encuadre. Un solo modelo cubre texto a imagen, edición con imágenes de referencia y diseño con cajas. La descripción que genera se devuelve como salida de <code>prompt</code>. En el ejemplo anterior, solicítela junto con las imágenes para ver lo que recibió el modelo de imagen:</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-python "><!---->out = pipe(
    prompt=<span class="hljs-string">&quot;A red fox in a snow-covered forest clearing at golden hour&quot;</span>,
    generator=torch.Generator().manual_seed(<span class="hljs-number">0</span>),
    output=[<span class="hljs-string">&quot;images&quot;</span>, <span class="hljs-string">&quot;prompt&quot;</span>],
)
<span class="hljs-built_in">print</span>(out[<span class="hljs-string">&quot;prompt&quot;</span>])<!----></pre></div><!----> <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=" "><!---->Full-color outdoor wildlife photograph taken <span class="hljs-built_in">from</span> <span class="hljs-keyword">a</span> low eye-level perspective <span class="hljs-keyword">in</span> <span class="hljs-keyword">a</span> snow-covered forest clearing <span class="hljs-keyword">at</span> golden hour. In <span class="hljs-keyword">the</span> center foreground, <span class="hljs-keyword">a</span> red fox stands <span class="hljs-keyword">in</span> profile facing left, its body turned toward <span class="hljs-keyword">the</span> viewer <span class="hljs-keyword">and</span> its head raised <span class="hljs-keyword">as</span> <span class="hljs-keyword">it</span> looks across <span class="hljs-keyword">the</span> clearing. The fox has dense rust-red winter fur, <span class="hljs-keyword">a</span> white throat <span class="hljs-keyword">and</span> chest, <span class="hljs-keyword">black</span> <span class="hljs-built_in">lower</span> legs, pointed ears <span class="hljs-keyword">with</span> dark tips, <span class="hljs-keyword">and</span> <span class="hljs-keyword">a</span> <span class="hljs-keyword">long</span> bushy tail ending <span class="hljs-keyword">in</span> white. ...<!----></pre></div><!----> <p>La leyenda completa tiene aproximadamente 240 palabras; <code>max_new_tokens</code> limita su longitud (3584 por defecto).</p> <p>Puedes imprimir el <code>doc</code> del paso para ver qué recibe y qué escribe:</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-python "><!----><span class="hljs-built_in">print</span>(pipe.blocks.sub_blocks[<span class="hljs-string">&quot;prompt_upsampler&quot;</span>].doc)<!----></pre></div><!----> <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=" "><!---->  <span class="hljs-symbol">Inputs</span>:
      prompt (<span class="hljs-string">`str`</span>):
          <span class="hljs-symbol">The</span> edit instruction to expand into the full caption (one per call).
      image (<span class="hljs-string">`Image | list`</span>, *optional*):
          <span class="hljs-symbol">The</span> reference image(s) of the edit, at most <span class="hljs-number">10</span>; the same ones the pipeline receives.
      boxes (<span class="hljs-string">`list`</span>, *optional*):
          <span class="hljs-symbol">Layout</span> boxes, each {<span class="hljs-string">&#x27;id&#x27;</span>, <span class="hljs-string">&#x27;desc&#x27;</span>, <span class="hljs-string">&#x27;from&#x27;</span>, <span class="hljs-string">&#x27;src_bbox&#x27;</span>, <span class="hljs-string">&#x27;tgt_bbox&#x27;</span>} with the boxes as [y_min, x_min, y_max, x_max] in
          <span class="hljs-number">0</span><span class="hljs-number">-1</span>: ...
      max_new_tokens (<span class="hljs-string">`int`</span>, *optional*, defaults to <span class="hljs-number">3584</span>):
          <span class="hljs-symbol">The</span> most tokens the upsampler may generate for the caption.

  <span class="hljs-symbol">Outputs</span>:
      prompt (<span class="hljs-string">`str`</span>):
          <span class="hljs-symbol">The</span> full caption, replacing the request.<!----></pre></div><!----> <p>Como otros bloques de la pipeline, el bloque <code>prompt_upsampler</code> también se ejecuta de forma independiente. La forma recomendada es extraerlo de los bloques: el bloque extraído se convierte en un amplificador independiente, y los bloques restantes forman una pipeline que toma la descripción ampliada como <code>prompt</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-python "><!---->blocks = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>).blocks
pup_block = blocks.sub_blocks.pop(<span class="hljs-string">&quot;prompt_upsampler&quot;</span>)

pup = pup_block.init_pipeline(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>)
pup.load_components(dtype=torch.bfloat16)
pup.to(<span class="hljs-string">&quot;cuda&quot;</span>)
caption = pup(prompt=<span class="hljs-string">&quot;A red fox in a snow-covered forest clearing at golden hour&quot;</span>, output=<span class="hljs-string">&quot;prompt&quot;</span>)

pipe = blocks.init_pipeline(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)
image = pipe(prompt=caption, generator=torch.Generator().manual_seed(<span class="hljs-number">0</span>), output=<span class="hljs-string">&quot;images&quot;</span>)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <p>De esta forma, puede servirse como una etapa independiente, en su propia GPU o detrás de su propio endpoint, y una interfaz de usuario puede mostrar la descripción que escribió y permitir al usuario editarla antes de generar. Cualquier otro LLM puede ocupar el lugar del amplificador, con los mismos prompts de sistema: proporciónale <code>system_prompts["t2i"]</code> del repositorio del amplificador (o <code>"i2i"</code>, con las referencias primero en el mensaje del usuario) como prompt de sistema y la solicitud como mensaje del usuario, y pasa la descripción que devuelve como <code>prompt</code> a <code>pipe</code>.</p> <!--[1--><h2 class="relative group"><a id="image-to-image" 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="#image-to-image"><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><!---->Imagen a imagen<!----></span></h2><!--]--><!----> <p>Pase una o más imágenes de referencia con <code>image</code>. Una sola referencia se mantiene íntegra: solo se reduce si supera los 4 megapíxeles, se rellena hasta la cuadrícula de 16 píxeles y la generación usa su tamaño por defecto, por lo que la edición se devuelve con el tamaño de la entrada. Varias referencias se ajustan cada una a menos de 1 megapíxel y se recortan a la cuadrícula, y la generación usa por defecto el tamaño de la primera. Un <code>height</code>/<code>width</code> dado anula el valor por defecto. Describa la imagen <em>objetivo</em> como una leyenda completa de la referencia con su edición integrada, en lugar de una instrucción de una línea, o deje que el <a href="#prompt-upsampling">prompt upsampler</a> la escriba a partir de la instrucción y las referencias:</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-python "><!---->night = pipe(
    prompt=<span class="hljs-string">&quot;the same scene at night under a full moon&quot;</span>,
    image=image,
    generator=torch.Generator().manual_seed(<span class="hljs-number">0</span>),
    output=[<span class="hljs-string">&quot;images&quot;</span>, <span class="hljs-string">&quot;prompt&quot;</span>],
)<!----></pre></div><!----> <p>Con varias referencias, hágase referencia a ellas como la primera imagen, la segunda imagen, … en el orden en que se pasan a <code>image</code>; la descripción las denomina <code>&lt;ref_image_0></code>, <code>&lt;ref_image_1></code>, … para el modelo de imagen:</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-python "><!---->combined = pipe(
    prompt=<span class="hljs-string">&quot;the fox from the first image sitting on the wooden bench from the second image&quot;</span>,
    image=[image, bench],
    height=<span class="hljs-number">768</span>,
    width=<span class="hljs-number">1360</span>,
    output=<span class="hljs-string">&quot;images&quot;</span>,
)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <p>Pasa <code>height</code> y/o <code>width</code> para anular el tamaño derivado.</p> <!--[1--><h2 class="relative group"><a id="layout-control" 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="#layout-control"><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><!---->Control de diseño<!----></span></h2><!--]--><!----> <p>To place content explicitly, pass <code>boxes</code>: for each element an <code>id</code>, a short <code>desc</code>, and a <code>bbox</code> of <code>[y_min, x_min, y_max, x_max]</code> in 0-1 (origin top-left), so it is independent of the output size. The upsampler writes the caption around them, citing each element as <code>&lt;id></code>, and appends the table of boxes, yours kept exactly:</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-python "><!---->boxes = [
    {<span class="hljs-string">&quot;id&quot;</span>: <span class="hljs-string">&quot;balloon_1&quot;</span>, <span class="hljs-string">&quot;desc&quot;</span>: <span class="hljs-string">&quot;a red and yellow striped hot-air balloon&quot;</span>, <span class="hljs-string">&quot;bbox&quot;</span>: [<span class="hljs-number">0.08</span>, <span class="hljs-number">0.06</span>, <span class="hljs-number">0.48</span>, <span class="hljs-number">0.38</span>]},
    {<span class="hljs-string">&quot;id&quot;</span>: <span class="hljs-string">&quot;balloon_2&quot;</span>, <span class="hljs-string">&quot;desc&quot;</span>: <span class="hljs-string">&quot;a small blue hot-air balloon&quot;</span>, <span class="hljs-string">&quot;bbox&quot;</span>: [<span class="hljs-number">0.20</span>, <span class="hljs-number">0.62</span>, <span class="hljs-number">0.42</span>, <span class="hljs-number">0.80</span>]},
]
out = pipe(prompt=<span class="hljs-string">&quot;hot-air balloons over a green valley at sunrise&quot;</span>, boxes=boxes, output=[<span class="hljs-string">&quot;images&quot;</span>, <span class="hljs-string">&quot;prompt&quot;</span>])<!----></pre></div><!----> <p>Al editar, una caja indica su origen con <code>from</code>: <code>"target"</code> con un <code>tgt_bbox</code> añade un nuevo elemento allí, <code>"ref_image_0"</code> con un <code>src_bbox</code> y un <code>tgt_bbox</code> mueve un elemento de esa referencia (las cajas iguales lo mantienen en su lugar):</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-python "><!---->boxes = [{<span class="hljs-string">&quot;id&quot;</span>: <span class="hljs-string">&quot;cactus&quot;</span>, <span class="hljs-string">&quot;desc&quot;</span>: <span class="hljs-string">&quot;a small potted cactus&quot;</span>, <span class="hljs-string">&quot;from&quot;</span>: <span class="hljs-string">&quot;target&quot;</span>, <span class="hljs-string">&quot;src_bbox&quot;</span>: <span class="hljs-literal">None</span>, <span class="hljs-string">&quot;tgt_bbox&quot;</span>: [<span class="hljs-number">0.55</span>, <span class="hljs-number">0.62</span>, <span class="hljs-number">0.85</span>, <span class="hljs-number">0.85</span>]}]
out = pipe(prompt=<span class="hljs-string">&quot;add a small potted cactus in the marked spot&quot;</span>, image=mug, boxes=boxes, output=[<span class="hljs-string">&quot;images&quot;</span>, <span class="hljs-string">&quot;prompt&quot;</span>])

boxes = [{<span class="hljs-string">&quot;id&quot;</span>: <span class="hljs-string">&quot;knight&quot;</span>, <span class="hljs-string">&quot;desc&quot;</span>: <span class="hljs-string">&quot;the knight figure&quot;</span>, <span class="hljs-string">&quot;from&quot;</span>: <span class="hljs-string">&quot;ref_image_0&quot;</span>, <span class="hljs-string">&quot;src_bbox&quot;</span>: [<span class="hljs-number">0.50</span>, <span class="hljs-number">0.15</span>, <span class="hljs-number">0.85</span>, <span class="hljs-number">0.35</span>], <span class="hljs-string">&quot;tgt_bbox&quot;</span>: [<span class="hljs-number">0.19</span>, <span class="hljs-number">0.06</span>, <span class="hljs-number">0.54</span>, <span class="hljs-number">0.26</span>]}]
out = pipe(prompt=<span class="hljs-string">&quot;move the knight up and to the left, keep the rest unchanged&quot;</span>, image=knight, boxes=boxes, output=[<span class="hljs-string">&quot;images&quot;</span>, <span class="hljs-string">&quot;prompt&quot;</span>])<!----></pre></div><!----> <p>La leyenda que lee el modelo de imagen tiene los elementos etiquetados con <code>&lt;id></code> en el texto y la tabla añadida como JSON, con las cajas en un lienzo de 0 a 1000; imprime la salida de <code>prompt</code> para verla. Una leyenda escrita a mano en ese formato se envía a la pipeline sin ese paso, como en <a href="#prompt-upsampling">Prompt upsampling</a>.</p> <!--[1--><h2 class="relative group"><a id="guidance-scale" 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="#guidance-scale"><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><!---->Escala de guía<!----></span></h2><!--]--><!----> <p>El pipeline aplica la guía sin clasificador a través de su componente <code>guider</code> y no acepta un argumento <code>guidance_scale</code> en el momento de la llamada. El guider predeterminado es <a href="/docs/diffusers/main/es/api/modular_diffusers/guiders#diffusers.ClassifierFreeGuidance">ClassifierFreeGuidance</a> con <code>guidance_scale=3.5</code>, la configuración recomendada para ambos flujos de trabajo. Para cambiarlo, cree una copia actualizada del guider:</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-python "><!---->pipe.guider = pipe.guider.new(guidance_scale=<span class="hljs-number">5.0</span>)<!----></pre></div><!----> <p>Consulte la <a href="../../using-diffusers/guiders">guía de guías</a> para cambiar a un método de guía diferente.</p> <!--[1--><h2 class="relative group"><a id="step-distilled-model" 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="#step-distilled-model"><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><!---->Modelo destilado por pasos<!----></span></h2><!--]--><!----> <p>El checkpoint destilado por pasos genera en 16 pasos sin guía libre de clasificador. Su transformer es un mapa de flujo: en cada paso se le indica el paso temporal actual y el paso al que salta. Se carga como el modelo base (el <code>modular_model_index.json</code> del repositorio se resuelve en <a href="/docs/diffusers/main/es/api/pipelines/flux3#diffusers.Flux3DistilledModularPipeline">Flux3DistilledModularPipeline</a> y <a href="/docs/diffusers/main/es/api/pipelines/flux3#diffusers.Flux3DistilledAutoBlocks">Flux3DistilledAutoBlocks</a>); el pipeline no tiene <code>guider</code> y no acepta <code>negative_prompt</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-python "><!---->pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein&quot;</span>)
pipe.load_components(dtype=torch.bfloat16)
pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)

image = pipe(prompt=prompt, generator=torch.Generator().manual_seed(<span class="hljs-number">0</span>), output=<span class="hljs-string">&quot;images&quot;</span>)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <p>Sirve para los mismos dos flujos de trabajo que el modelo base. <code>num_inference_steps</code> tiene un valor predeterminado de 16; el modelo se entrena para 4 a 16 pasos.</p> <!--[1--><h2 class="relative group"><a id="reproducible-generation" 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="#reproducible-generation"><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><!---->Generación reproducible<!----></span></h2><!--]--><!----> <p>Un <code>generator</code> fija el ruido que extrae el pipeline (consulte <a href="../../using-diffusers/reusing_seeds">Reproducibilidad</a>). Para reutilizar un ruido extraído en otro lugar, páselo como <code>latents</code> con forma <code>(batch_size, 128, height // 16, width // 16)</code>. Por ejemplo, la <a href="https://github.com/black-forest-labs/flux-3-image-klein" rel="nofollow">implementación de BFL</a> extrae su ruido como tokens de imagen en la CPU y lo convierte a bf16; esto reproduce su muestra para una semilla dada:</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-python "><!---->height, width = <span class="hljs-number">1024</span>, <span class="hljs-number">1024</span>
h, w = height // <span class="hljs-number">16</span>, width // <span class="hljs-number">16</span>

noise = torch.randn((h * w, <span class="hljs-number">128</span>), generator=torch.Generator().manual_seed(<span class="hljs-number">0</span>)).to(torch.bfloat16)
latents = noise.T.reshape(<span class="hljs-number">1</span>, <span class="hljs-number">128</span>, h, w)

image = pipe(prompt=prompt, latents=latents, height=height, width=width, output=<span class="hljs-string">&quot;images&quot;</span>)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="workflows" 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="#workflows"><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><!---->Flujos de trabajo<!----></span></h2><!--]--><!----> <p>El pipeline atiende dos flujos de trabajo, <code>text2image</code> y <code>image_conditioned</code>; el que se ejecuta se elige según las entradas (<code>image</code> selecciona image-to-image). <a href="/docs/diffusers/main/es/api/modular_diffusers/pipeline#diffusers.ModularPipeline.from_pretrained">from_pretrained()</a> también puede construir un pipeline para un solo flujo de trabajo con el argumento <code>workflow</code>, que entonces solo espera y carga los componentes que ese flujo de trabajo necesita:</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-python "><!----><span class="hljs-keyword">from</span> diffusers.modular_pipelines <span class="hljs-keyword">import</span> Flux3AutoBlocks

<span class="hljs-built_in">print</span>(Flux3AutoBlocks().available_workflows)
<span class="hljs-comment"># [&quot;text2image&quot;, &quot;image_conditioned&quot;]</span>

t2i = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>, workflow=<span class="hljs-string">&quot;text2image&quot;</span>)
<span class="hljs-built_in">print</span>(t2i.blocks.doc)<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="per-step-callbacks" 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="#per-step-callbacks"><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><!---->Callbacks por paso<!----></span></h2><!--]--><!----> <p>Para usar una función de callback existente, inserte un bloque en el bucle de desruido. La <a href="../../using-diffusers/callback#modular-pipelines">guía de callbacks</a> tiene un ejemplo sencillo con el bloque <code>CallbackOnStepEnd</code> y el helper <code>walk_denoise_loops</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-python "><!---->blocks = Flux3AutoBlocks()  <span class="hljs-comment"># or Flux3DistilledAutoBlocks</span>
<span class="hljs-keyword">for</span> loop <span class="hljs-keyword">in</span> walk_denoise_loops(blocks):
    loop.sub_blocks[<span class="hljs-string">&quot;callback&quot;</span>] = CallbackOnStepEnd()
pipe = blocks.init_pipeline(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>)

image = pipe(prompt=caption, callback_on_step_end=my_callback, output=<span class="hljs-string">&quot;images&quot;</span>)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <p>Para detener el proceso de forma anticipada, lanza una excepción desde el callback; no existe una bandera <code>interrupt</code>.</p> <!--[1--><h2 class="relative group"><a id="memory" 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="#memory"><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><!---->Memoria<!----></span></h2><!--]--><!----> <p>Cargar todo en bf16 requiere aproximadamente 39 GB de pesos: el transformer tiene 5.5B de parámetros, el codificador de texto Qwen3-VL-4B tiene 4B y el <a href="#prompt-upsampling">prompt upsampler</a> tiene 9B (unos 19 GB por sí solo), con el VAE de imagen en fp32. Un pipeline construido sin el paso de upsampler (ver allí) carga aproximadamente 20 GB. Para GPUs con menos memoria, habilite el offloading de componentes; el gestor mueve los componentes entre la CPU y la GPU según lo que necesite cada paso, incluido el upsampler:</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-python "><!----><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> ComponentsManager, ModularPipeline

manager = ComponentsManager()
pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>, components_manager=manager)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
manager.enable_auto_cpu_offload(device=<span class="hljs-string">&quot;cuda&quot;</span>)<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="single-file-loading" 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="#single-file-loading"><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 un solo archivo<!----></span></h2><!--]--><!----> <p><a href="/docs/diffusers/main/es/api/models/flux3_transformer#diffusers.Flux3Transformer2DModel">Flux3Transformer2DModel</a> carga el transformer directamente desde un paquete de modelos de Black Forest Labs con <a href="/docs/diffusers/main/es/api/loaders/single_file#diffusers.loaders.FromOriginalModelMixin.from_single_file">from_single_file()</a>. El paquete <code>config.json</code> no se lee; pase la configuración de diffusers con <code>config</code> y <code>subfolder</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-python "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> Flux3Transformer2DModel, ModularPipeline

transformer = Flux3Transformer2DModel.from_single_file(
    <span class="hljs-string">&quot;https://huggingface.co/black-forest-labs/flux-3-image-klein-base/blob/main/model.safetensors&quot;</span>,
    config=<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>,
    subfolder=<span class="hljs-string">&quot;transformer&quot;</span>,
    dtype=torch.bfloat16,
)

pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
pipe.update_components(transformer=transformer)
pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)<!----></pre></div><!----> <p>Cargue primero los demás componentes por nombre: <a href="/docs/diffusers/main/es/api/modular_diffusers/pipeline#diffusers.ModularPipeline.load_components">load_components()</a> sin <code>names</code> también cargaría el transformer del repositorio de la pipeline.</p> <p>La carga de los puntos de control de archivo único FP8r y NVFP4 también está admitida. Ambos se cargan mediante la misma llamada con una <a href="/docs/diffusers/main/es/api/quantization#diffusers.BFLQuantizationConfig">BFLQuantizationConfig</a>. Consulte la <a href="../../quantization/bfl">guía de cuantización de BFL</a> para conocer los requisitos de kernel de cada esquema.</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-python "><!----><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> BFLQuantizationConfig

transformer = Flux3Transformer2DModel.from_single_file(
    <span class="hljs-string">&quot;https://huggingface.co/black-forest-labs/flux-3-image-klein-base/blob/main/variants/nvfp4/model.safetensors&quot;</span>,
    config=<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein-base&quot;</span>,
    subfolder=<span class="hljs-string">&quot;transformer&quot;</span>,
    quantization_config=BFLQuantizationConfig(compute_dtype=torch.bfloat16),
    dtype=torch.bfloat16,
)<!----></pre></div><!----> <p>Al cargar los puntos de control destilados por pasos, necesitamos sus
configuraciones (<code>flow_map=True</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-python "><!---->transformer = Flux3Transformer2DModel.from_single_file(
    <span class="hljs-string">&quot;https://huggingface.co/black-forest-labs/flux-3-image-klein/blob/main/variants/nvfp4/model.safetensors&quot;</span>,
    config=<span class="hljs-string">&quot;black-forest-labs/flux-3-image-klein&quot;</span>,
    subfolder=<span class="hljs-string">&quot;transformer&quot;</span>,
    quantization_config=BFLQuantizationConfig(compute_dtype=torch.bfloat16),
    dtype=torch.bfloat16,
)<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="diffusers.Flux3ModularPipeline" 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="#diffusers.Flux3ModularPipeline"><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><!---->Flux3ModularPipeline<!----></span></h2><!--]--><!----> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><div><span class="group flex space-x-1.5 items-center text-gray-800 bg-gradient-to-r rounded-tr-lg -mt-4 -ml-4 pt-3 px-2.5" id="diffusers.Flux3ModularPipeline"><!----><h3 class="!m-0"><span class="flex-1 break-all md:text-lg bg-gradient-to-r px-2.5 py-1.5 rounded-xl from-indigo-50/70 to-white dark:from-gray-900 dark:to-gray-950 dark:text-indigo-300 text-indigo-700"><svg class="mr-1.5 text-indigo-500 inline-block -mt-0.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width=".8em" height=".8em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 24 24"><path class="uim-quaternary" d="M20.23 7.24L12 12L3.77 7.24a1.98 1.98 0 0 1 .7-.71L11 2.76c.62-.35 1.38-.35 2 0l6.53 3.77c.29.173.531.418.7.71z" opacity=".25" fill="currentColor"></path><path class="uim-tertiary" d="M12 12v9.5a2.09 2.09 0 0 1-.91-.21L4.5 17.48a2.003 2.003 0 0 1-1-1.73v-7.5a2.06 2.06 0 0 1 .27-1.01L12 12z" opacity=".5" fill="currentColor"></path><path class="uim-primary" d="M20.5 8.25v7.5a2.003 2.003 0 0 1-1 1.73l-6.62 3.82c-.275.13-.576.198-.88.2V12l8.23-4.76c.175.308.268.656.27 1.01z" fill="currentColor"></path></svg><span class="font-light">clase</span> <span class="font-medium">diffusers.</span><span class="font-semibold">Flux3ModularPipeline</span></span></h3><!----> <a id="diffusers.Flux3ModularPipeline" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.Flux3ModularPipeline"><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><!----></a> <!--[0--><a class="!ml-auto !text-gray-400 !no-underline text-sm flex items-center" href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/modular_pipelines/flux3/modular_pipeline.py#L24" target="_blank"><span>&lt;</span> <span class="hidden md:block mx-0.5 hover:!underline">source</span> <span>></span></a><!--]--></span> <!--[0--><p class="font-mono text-xs md:text-sm !leading-relaxed !my-6"><span>(</span> <!--[--><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">blocks<span class="opacity-60">: diffusers.modular_pipelines.modular_pipeline.ModularPipelineBlocks | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">pretrained_model_name_or_path<span class="opacity-60">: str | os.PathLike | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">components_manager<span class="opacity-60">: diffusers.modular_pipelines.components_manager.ComponentsManager | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">collection<span class="opacity-60">: str | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">workflow<span class="opacity-60">: str | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">modular_config_dict<span class="opacity-60">: dict[str, typing.Any] | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">config_dict<span class="opacity-60">: dict[str, typing.Any] | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">**kwargs<span class="opacity-60"></span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Un ModularPipeline para FLUX.3.</p></div> <!--[1--><h2 class="relative group"><a id="diffusers.Flux3AutoBlocks" 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="#diffusers.Flux3AutoBlocks"><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><!---->Flux3AutoBlocks<!----></span></h2><!--]--><!----> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><div><span class="group flex space-x-1.5 items-center text-gray-800 bg-gradient-to-r rounded-tr-lg -mt-4 -ml-4 pt-3 px-2.5" id="diffusers.Flux3AutoBlocks"><!----><h3 class="!m-0"><span class="flex-1 break-all md:text-lg bg-gradient-to-r px-2.5 py-1.5 rounded-xl from-indigo-50/70 to-white dark:from-gray-900 dark:to-gray-950 dark:text-indigo-300 text-indigo-700"><svg class="mr-1.5 text-indigo-500 inline-block -mt-0.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width=".8em" height=".8em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 24 24"><path class="uim-quaternary" d="M20.23 7.24L12 12L3.77 7.24a1.98 1.98 0 0 1 .7-.71L11 2.76c.62-.35 1.38-.35 2 0l6.53 3.77c.29.173.531.418.7.71z" opacity=".25" fill="currentColor"></path><path class="uim-tertiary" d="M12 12v9.5a2.09 2.09 0 0 1-.91-.21L4.5 17.48a2.003 2.003 0 0 1-1-1.73v-7.5a2.06 2.06 0 0 1 .27-1.01L12 12z" opacity=".5" fill="currentColor"></path><path class="uim-primary" d="M20.5 8.25v7.5a2.003 2.003 0 0 1-1 1.73l-6.62 3.82c-.275.13-.576.198-.88.2V12l8.23-4.76c.175.308.268.656.27 1.01z" fill="currentColor"></path></svg><span class="font-light">clase</span> <span class="font-medium">diffusers.</span><span class="font-semibold">Flux3AutoBlocks</span></span></h3><!----> <a id="diffusers.Flux3AutoBlocks" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.Flux3AutoBlocks"><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><!----></a> <!--[0--><a class="!ml-auto !text-gray-400 !no-underline text-sm flex items-center" href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/modular_pipelines/flux3/modular_blocks_flux3.py#L339" target="_blank"><span>&lt;</span> <span class="hidden md:block mx-0.5 hover:!underline">source</span> <span>></span></a><!--]--></span> <!--[0--><p class="font-mono text-xs md:text-sm !leading-relaxed !my-6"><span>(</span> <!--[--><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Auto bloques que realizan la generación de texto a imagen y de imagen a imagen usando FLUX.3. <code>prompt</code> pasa primero por el amplificador de prompts; elimina el paso <code>prompt_upsampler</code> para codificarlo tal como se proporciona.</p> <p>Flujos de trabajo admitidos:</p> <ul><li><code>text2image</code>: requiere <code>prompt</code></li> <li><code>image_conditioned</code>: requiere <code>prompt</code>, <code>image</code></li></ul> <p>Componentes:
prompt_upsampler (<code>Qwen3_5ForConditionalGeneration</code>): El amplificador de prompts. prompt_upsampler_processor
(<code>AutoProcessor</code>): Su procesador. text_encoder (<code>Qwen3VLForConditionalGeneration</code>) tokenizer
(<code>AutoProcessor</code>) image_processor (<code>Flux3ImageProcessor</code>) vae (<code>AutoencoderKLFlux2</code>) transformer
(<code>Flux3Transformer2DModel</code>) scheduler (<code>FlowMatchEulerDiscreteScheduler</code>) guider (<code>ClassifierFreeGuidance</code>)</p> <p>Entradas:
prompt (<code>str</code>):
La instrucción de edición para expandir en la descripción completa (una por llamada).
image (<code>Image | list</code>, <em>opcional</em>):
La(s) imagen(es) de referencia de la edición, como máximo 10; las mismas que recibe el pipeline.
boxes (<code>list</code>, <em>opcional</em>):
Cajas de diseño, cada una {‘id’, ‘desc’, ‘from’, ‘src_bbox’, ‘tgt_bbox’} con las cajas como [y_min, x_min, y_max,
x_max] en 0-1: ‘from’: ‘target’ con <code>tgt_bbox</code> añade un elemento allí, ‘from’: ‘ref_image_0’, … con <code>src_bbox</code> y <code>tgt_bbox</code> mueve uno desde una imagen de referencia.
max_new_tokens (<code>int</code>, <em>opcional</em>, predeterminado a 3584):
El número máximo de tokens que el amplificador puede generar para la descripción.
negative_prompt (<code>str</code>, <em>opcional</em>):
El prompt que no debe guiar la generación
max_sequence_length (<code>int</code>, <em>opcional</em>, predeterminado a 8192):
Número máximo de tokens del prompt; los prompts más largos se truncan.
text_encoder_out_layers (<code>tuple</code>, <em>opcional</em>, predeterminado a (4, 8, 12, 16, 20, 24, 28, 32)):
Índices de las capas ocultas del codificador de texto cuyos estados se apilan para formar los embeddings de texto
num_images_per_prompt (<code>int</code>, <em>opcional</em>, predeterminado a 1):
El número de imágenes a generar por prompt.
image_cond_latents (<code>list</code>, <em>opcional</em>):
Latentes normalizados y empaquetados de las imágenes de referencia, un tensor de forma (B, C, height_i//16, width_i//16)
por imagen
height (<code>int</code>, <em>opcional</em>, predeterminado a None o 1024, dependiendo del flujo de trabajo):
La altura en píxeles de la imagen generada; predeterminado a la altura propia de la primera imagen de referencia
width (<code>int</code>, <em>opcional</em>, predeterminado a None o 1024, dependiendo del flujo de trabajo):
El ancho en píxeles de la imagen generada; predeterminado al ancho propio de la primera imagen de referencia
reference_crop (<code>list | NoneType</code>, <em>opcional</em>):
Por referencia, la caja de la referencia dentro de su lienzo relleno o None, del paso de preprocesamiento
latents (<code>Tensor</code>):
Latentes ruidosos pregenerados para la generación de imágenes.
generator (<code>Generator</code>, <em>opcional</em>):
Generador de Torch para generación determinista.
num_inference_steps (<code>int</code>):
El número de pasos de desruido.
timesteps (<code>Tensor</code>):
Pasos de tiempo para el proceso de desruido. (anula <code>num_inference_steps</code> cuando se proporciona)
sigmas (<code>list</code>, <em>opcional</em>):
Sigmas personalizados para el proceso de desruido. (anula <code>num_inference_steps</code> cuando se proporciona)
*<em>denoiser_input_fields (<code>None</code>,</em>opcional<em>):
entradas condicionales del modelo para el desruidador: p. ej. prompt_embeds, negative_prompt_embeds, etc.
attention_kwargs (<code>dict</code>,</em>opcional<em>):
Kwargs adicionales para los procesadores de atención.
output_type (<code>str</code>,</em>opcional*, predeterminado a pil):
Formato de salida: ‘pil’, ‘np’, ‘pt’.</p> <p>Salidas:
images (<code>list</code>):
Imágenes generadas.
latents (<code>Tensor</code>):
Latentes desruidados.</p></div> <!--[1--><h2 class="relative group"><a id="diffusers.Flux3DistilledAutoBlocks" 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="#diffusers.Flux3DistilledAutoBlocks"><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><!---->Flux3DistilledAutoBlocks<!----></span></h2><!--]--><!----> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><div><span class="group flex space-x-1.5 items-center text-gray-800 bg-gradient-to-r rounded-tr-lg -mt-4 -ml-4 pt-3 px-2.5" id="diffusers.Flux3DistilledAutoBlocks"><!----><h3 class="!m-0"><span class="flex-1 break-all md:text-lg bg-gradient-to-r px-2.5 py-1.5 rounded-xl from-indigo-50/70 to-white dark:from-gray-900 dark:to-gray-950 dark:text-indigo-300 text-indigo-700"><svg class="mr-1.5 text-indigo-500 inline-block -mt-0.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width=".8em" height=".8em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 24 24"><path class="uim-quaternary" d="M20.23 7.24L12 12L3.77 7.24a1.98 1.98 0 0 1 .7-.71L11 2.76c.62-.35 1.38-.35 2 0l6.53 3.77c.29.173.531.418.7.71z" opacity=".25" fill="currentColor"></path><path class="uim-tertiary" d="M12 12v9.5a2.09 2.09 0 0 1-.91-.21L4.5 17.48a2.003 2.003 0 0 1-1-1.73v-7.5a2.06 2.06 0 0 1 .27-1.01L12 12z" opacity=".5" fill="currentColor"></path><path class="uim-primary" d="M20.5 8.25v7.5a2.003 2.003 0 0 1-1 1.73l-6.62 3.82c-.275.13-.576.198-.88.2V12l8.23-4.76c.175.308.268.656.27 1.01z" fill="currentColor"></path></svg><span class="font-light">clase</span> <span class="font-medium">diffusers.</span><span class="font-semibold">Flux3DistilledAutoBlocks</span></span></h3><!----> <a id="diffusers.Flux3DistilledAutoBlocks" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.Flux3DistilledAutoBlocks"><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><!----></a> <!--[0--><a class="!ml-auto !text-gray-400 !no-underline text-sm flex items-center" href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/modular_pipelines/flux3/modular_blocks_flux3_distilled.py#L258" target="_blank"><span>&lt;</span> <span class="hidden md:block mx-0.5 hover:!underline">source</span> <span>></span></a><!--]--></span> <!--[0--><p class="font-mono text-xs md:text-sm !leading-relaxed !my-6"><span>(</span> <!--[--><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Auto bloques que realizan la generación de texto a imagen y de imagen a imagen utilizando el modelo FLUX.3 destilado por pasos (16 pasos, sin guía libre de clasificador). <code>prompt</code> pasa primero por el amplificador de prompts; elimina el paso <code>prompt_upsampler</code> para codificarlo tal como se proporciona.</p> <p>Flujos de trabajo admitidos:</p> <ul><li><code>text2image</code>: requiere <code>prompt</code></li> <li><code>image_conditioned</code>: requiere <code>prompt</code>, <code>image</code></li></ul> <p>Componentes:
prompt_upsampler (<code>Qwen3_5ForConditionalGeneration</code>): El amplificador de prompts. prompt_upsampler_processor
(<code>AutoProcessor</code>): Su procesador. text_encoder (<code>Qwen3VLForConditionalGeneration</code>) tokenizer
(<code>AutoProcessor</code>) image_processor (<code>Flux3ImageProcessor</code>) vae (<code>AutoencoderKLFlux2</code>) transformer
(<code>Flux3Transformer2DModel</code>) scheduler (<code>FlowMatchEulerDiscreteScheduler</code>)</p> <p>Entradas:
prompt (<code>str</code>):
La instrucción de edición para expandir en la descripción completa (una por llamada).
image (<code>Image | list</code>, <em>opcional</em>):
La(s) imagen(es) de referencia de la edición, como máximo 10; las mismas que recibe el pipeline.
boxes (<code>list</code>, <em>opcional</em>):
Cajas de diseño, cada una {‘id’, ‘desc’, ‘from’, ‘src_bbox’, ‘tgt_bbox’} con las cajas como [y_min, x_min, y_max,
x_max] en 0-1: ‘from’: ‘target’ con <code>tgt_bbox</code> añade un elemento allí, ‘from’: ‘ref_image_0’, … con <code>src_bbox</code> y <code>tgt_bbox</code> mueve uno desde una imagen de referencia.
max_new_tokens (<code>int</code>, <em>opcional</em>, predeterminado a 3584):
El número máximo de tokens que el amplificador puede generar para la descripción.
max_sequence_length (<code>int</code>, <em>opcional</em>, predeterminado a 8192):
Número máximo de tokens del prompt; los prompts más largos se truncan.
text_encoder_out_layers (<code>tuple</code>, <em>opcional</em>, predeterminado a (4, 8, 12, 16, 20, 24, 28, 32)):
Índices de las capas ocultas del codificador de texto cuyos estados se apilan para formar los embeddings de texto
num_images_per_prompt (<code>int</code>, <em>opcional</em>, predeterminado a 1):
El número de imágenes a generar por prompt.
image_cond_latents (<code>list</code>, <em>opcional</em>):
Latentes empaquetados normalizados de las imágenes de referencia, un tensor de forma (B, C, height_i//16, width_i//16)
por imagen
height (<code>int</code>, <em>opcional</em>, predeterminado a None o 1024, dependiendo del flujo de trabajo):
La altura en píxeles de la imagen generada; predeterminado a la altura propia de la primera imagen de referencia
width (<code>int</code>, <em>opcional</em>, predeterminado a None o 1024, dependiendo del flujo de trabajo):
El ancho en píxeles de la imagen generada; predeterminado al ancho propio de la primera imagen de referencia
reference_crop (<code>list | NoneType</code>, <em>opcional</em>):
Por referencia, la caja de la referencia dentro de su lienzo relleno o None, del paso de preprocesamiento
latents (<code>Tensor</code>):
Latentes ruidosos pregenerados para la generación de imágenes.
generator (<code>Generator</code>, <em>opcional</em>):
Generador de Torch para generación determinista.
num_inference_steps (<code>int</code>):
El número de pasos de desruido.
timesteps (<code>Tensor</code>):
Pasos de tiempo para el proceso de desruido. (anula <code>num_inference_steps</code> cuando se proporciona)
sigmas (<code>list</code>, <em>opcional</em>):
Sigmas personalizados para el proceso de desruido. (anula <code>num_inference_steps</code> cuando se proporciona)
*<em>denoiser_input_fields (<code>None</code>,</em>opcional<em>):
entradas condicionales del modelo para el desruidador: p. ej. prompt_embeds, negative_prompt_embeds, etc.
output_type (<code>str</code>,</em>opcional*, predeterminado a pil):
Formato de salida: ‘pil’, ‘np’, ‘pt’.</p> <p>Salidas:
images (<code>list</code>):
Imágenes generadas.
latents (<code>Tensor</code>):
Latentes desruidados.</p></div> <!--[1--><h2 class="relative group"><a id="diffusers.Flux3DistilledModularPipeline" 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="#diffusers.Flux3DistilledModularPipeline"><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><!---->Flux3DistilledModularPipeline<!----></span></h2><!--]--><!----> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><div><span class="group flex space-x-1.5 items-center text-gray-800 bg-gradient-to-r rounded-tr-lg -mt-4 -ml-4 pt-3 px-2.5" id="diffusers.Flux3DistilledModularPipeline"><!----><h3 class="!m-0"><span class="flex-1 break-all md:text-lg bg-gradient-to-r px-2.5 py-1.5 rounded-xl from-indigo-50/70 to-white dark:from-gray-900 dark:to-gray-950 dark:text-indigo-300 text-indigo-700"><svg class="mr-1.5 text-indigo-500 inline-block -mt-0.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width=".8em" height=".8em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 24 24"><path class="uim-quaternary" d="M20.23 7.24L12 12L3.77 7.24a1.98 1.98 0 0 1 .7-.71L11 2.76c.62-.35 1.38-.35 2 0l6.53 3.77c.29.173.531.418.7.71z" opacity=".25" fill="currentColor"></path><path class="uim-tertiary" d="M12 12v9.5a2.09 2.09 0 0 1-.91-.21L4.5 17.48a2.003 2.003 0 0 1-1-1.73v-7.5a2.06 2.06 0 0 1 .27-1.01L12 12z" opacity=".5" fill="currentColor"></path><path class="uim-primary" d="M20.5 8.25v7.5a2.003 2.003 0 0 1-1 1.73l-6.62 3.82c-.275.13-.576.198-.88.2V12l8.23-4.76c.175.308.268.656.27 1.01z" fill="currentColor"></path></svg><span class="font-light">clase</span> <span class="font-medium">diffusers.</span><span class="font-semibold">Flux3DistilledModularPipeline</span></span></h3><!----> <a id="diffusers.Flux3DistilledModularPipeline" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.Flux3DistilledModularPipeline"><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><!----></a> <!--[0--><a class="!ml-auto !text-gray-400 !no-underline text-sm flex items-center" href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/modular_pipelines/flux3/modular_pipeline.py#L52" target="_blank"><span>&lt;</span> <span class="hidden md:block mx-0.5 hover:!underline">source</span> <span>></span></a><!--]--></span> <!--[0--><p class="font-mono text-xs md:text-sm !leading-relaxed !my-6"><span>(</span> <!--[--><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">blocks<span class="opacity-60">: diffusers.modular_pipelines.modular_pipeline.ModularPipelineBlocks | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">pretrained_model_name_or_path<span class="opacity-60">: str | os.PathLike | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">components_manager<span class="opacity-60">: diffusers.modular_pipelines.components_manager.ComponentsManager | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">collection<span class="opacity-60">: str | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">workflow<span class="opacity-60">: str | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">modular_config_dict<span class="opacity-60">: dict[str, typing.Any] | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">config_dict<span class="opacity-60">: dict[str, typing.Any] | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">**kwargs<span class="opacity-60"></span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Un ModularPipeline para el modelo FLUX.3 destilado por pasos (16 pasos, sin guía libre de clasificador).</p></div> <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/api/pipelines/flux3.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_gxjg5b = {
						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.U6TUchwJ.js"),
						import("/docs/diffusers/main/en/_app/immutable/entry/app.3qKhPoX-.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 8],
							data: [null,null],
							form: null,
							error: null
						});
					});
				}
			</script>
		
Xet Storage Details

Size:: 97.2 kB
Xet hash:: ba3edcb9256fd953afd6fb6fc0af625a18d54f5e28fecb85646656b2a4b602a4
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.