diffusers/pr_12595/en/_app/immutable/nodes/169.df53b9cf.js

import{s as Pn,o as xn,n as Kt}from"../chunks/scheduler.53228c21.js";import{S as In,i as jn,e as r,s as o,c,h as Sn,a as l,d as n,b as i,f as I,g as p,j as h,k as x,l as d,m as a,n as m,t as g,o as u,p as f}from"../chunks/index.100fac89.js";import{C as Zn}from"../chunks/CopyLLMTxtMenu.7aefc1a4.js";import{D as W}from"../chunks/Docstring.d6cb35e8.js";import{C as He}from"../chunks/CodeBlock.d30a6509.js";import{E as Ot}from"../chunks/ExampleCodeBlock.a12c1377.js";import{H as v,E as $n}from"../chunks/MermaidChart.svelte_svelte_type_style_lang.3722da43.js";function kn(G){let _,J="Examples:",w,b,y;return b=new He({props:{code:"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",highlighted:`<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">import</span> torch
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableCascadeCombinedPipeline

<span class="hljs-meta">&gt;&gt;&gt; </span>pipe = StableCascadeCombinedPipeline.from_pretrained(
<span class="hljs-meta">... </span>    <span class="hljs-string">&quot;stabilityai/stable-cascade&quot;</span>, variant=<span class="hljs-string">&quot;bf16&quot;</span>, torch_dtype=torch.bfloat16
<span class="hljs-meta">... </span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.enable_model_cpu_offload()
<span class="hljs-meta">&gt;&gt;&gt; </span>prompt = <span class="hljs-string">&quot;an image of a shiba inu, donning a spacesuit and helmet&quot;</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>images = pipe(prompt=prompt)`,wrap:!1}}),{c(){_=r("p"),_.textContent=J,w=o(),c(b.$$.fragment)},l(s){_=l(s,"P",{"data-svelte-h":!0}),h(_)!=="svelte-kvfsh7"&&(_.textContent=J),w=i(s),p(b.$$.fragment,s)},m(s,C){a(s,_,C),a(s,w,C),m(b,s,C),y=!0},p:Kt,i(s){y||(g(b.$$.fragment,s),y=!0)},o(s){u(b.$$.fragment,s),y=!1},d(s){s&&(n(_),n(w)),f(b,s)}}}function Wn(G){let _,J="Examples:",w,b,y;return b=new He({props:{code:"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",highlighted:`<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">import</span> torch
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableCascadePriorPipeline

<span class="hljs-meta">&gt;&gt;&gt; </span>prior_pipe = StableCascadePriorPipeline.from_pretrained(
<span class="hljs-meta">... </span>    <span class="hljs-string">&quot;stabilityai/stable-cascade-prior&quot;</span>, torch_dtype=torch.bfloat16
<span class="hljs-meta">... </span>).to(<span class="hljs-string">&quot;cuda&quot;</span>)

<span class="hljs-meta">&gt;&gt;&gt; </span>prompt = <span class="hljs-string">&quot;an image of a shiba inu, donning a spacesuit and helmet&quot;</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>prior_output = pipe(prompt)`,wrap:!1}}),{c(){_=r("p"),_.textContent=J,w=o(),c(b.$$.fragment)},l(s){_=l(s,"P",{"data-svelte-h":!0}),h(_)!=="svelte-kvfsh7"&&(_.textContent=J),w=i(s),p(b.$$.fragment,s)},m(s,C){a(s,_,C),a(s,w,C),m(b,s,C),y=!0},p:Kt,i(s){y||(g(b.$$.fragment,s),y=!0)},o(s){u(b.$$.fragment,s),y=!1},d(s){s&&(n(_),n(w)),f(b,s)}}}function Gn(G){let _,J="Examples:",w,b,y;return b=new He({props:{code:"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",highlighted:`<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">import</span> torch
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableCascadePriorPipeline, StableCascadeDecoderPipeline

<span class="hljs-meta">&gt;&gt;&gt; </span>prior_pipe = StableCascadePriorPipeline.from_pretrained(
<span class="hljs-meta">... </span>    <span class="hljs-string">&quot;stabilityai/stable-cascade-prior&quot;</span>, torch_dtype=torch.bfloat16
<span class="hljs-meta">... </span>).to(<span class="hljs-string">&quot;cuda&quot;</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>gen_pipe = StableCascadeDecoderPipeline.from_pretrain(
<span class="hljs-meta">... </span>    <span class="hljs-string">&quot;stabilityai/stable-cascade&quot;</span>, torch_dtype=torch.float16
<span class="hljs-meta">... </span>).to(<span class="hljs-string">&quot;cuda&quot;</span>)

<span class="hljs-meta">&gt;&gt;&gt; </span>prompt = <span class="hljs-string">&quot;an image of a shiba inu, donning a spacesuit and helmet&quot;</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>prior_output = pipe(prompt)
<span class="hljs-meta">&gt;&gt;&gt; </span>images = gen_pipe(prior_output.image_embeddings, prompt=prompt)`,wrap:!1}}),{c(){_=r("p"),_.textContent=J,w=o(),c(b.$$.fragment)},l(s){_=l(s,"P",{"data-svelte-h":!0}),h(_)!=="svelte-kvfsh7"&&(_.textContent=J),w=i(s),p(b.$$.fragment,s)},m(s,C){a(s,_,C),a(s,w,C),m(b,s,C),y=!0},p:Kt,i(s){y||(g(b.$$.fragment,s),y=!0)},o(s){u(b.$$.fragment,s),y=!1},d(s){s&&(n(_),n(w)),f(b,s)}}}function Bn(G){let _,J,w,b,y,s,C,Xe,H,en=`This model is built upon the <a href="https://openreview.net/forum?id=gU58d5QeGv" rel="nofollow">Würstchen</a> architecture and its main
difference to other models like Stable Diffusion is that it is working at a much smaller latent space. Why is this
important? The smaller the latent space, the <strong>faster</strong> you can run inference and the <strong>cheaper</strong> the training becomes.
How small is the latent space? Stable Diffusion uses a compression factor of 8, resulting in a 1024x1024 image being
encoded to 128x128. Stable Cascade achieves a compression factor of 42, meaning that it is possible to encode a
1024x1024 image to 24x24, while maintaining crisp reconstructions. The text-conditional model is then trained in the
highly compressed latent space. Previous versions of this architecture, achieved a 16x cost reduction over Stable
Diffusion 1.5.`,ze,Q,tn=`Therefore, this kind of model is well suited for usages where efficiency is important. Furthermore, all known extensions
like finetuning, LoRA, ControlNet, IP-Adapter, LCM etc. are possible with this method as well.`,Fe,X,nn='The original codebase can be found at <a href="https://github.com/Stability-AI/StableCascade" rel="nofollow">Stability-AI/StableCascade</a>.',qe,z,Ae,F,an=`Stable Cascade consists of three models: Stage A, Stage B and Stage C, representing a cascade to generate images,
hence the name “Stable Cascade”.`,Oe,q,on=`Stage A & B are used to compress images, similar to what the job of the VAE is in Stable Diffusion.
However, with this setup, a much higher compression of images can be achieved. While the Stable Diffusion models use a
spatial compression factor of 8, encoding an image with resolution of 1024 x 1024 to 128 x 128, Stable Cascade achieves
a compression factor of 42. This encodes a 1024 x 1024 image to 24 x 24, while being able to accurately decode the
image. This comes with the great benefit of cheaper training and inference. Furthermore, Stage C is responsible
for generating the small 24 x 24 latents given a text prompt.`,Ke,A,sn="The Stage C model operates on the small 24 x 24 latents and denoises the latents conditioned on text prompts. The model is also the largest component in the Cascade pipeline and is meant to be used with the <code>StableCascadePriorPipeline</code>",et,O,rn="The Stage B and Stage A models are used with the <code>StableCascadeDecoderPipeline</code> and are responsible for generating the final image given the small 24 x 24 latents.",tt,B,ln="<p>There are some restrictions on data types that can be used with the Stable Cascade models. The official checkpoints for the  <code>StableCascadePriorPipeline</code> do not support the <code>torch.float16</code> data type. Please use <code>torch.bfloat16</code> instead.</p> <p>In order to use the <code>torch.bfloat16</code> data type with the <code>StableCascadeDecoderPipeline</code> you need to have PyTorch 2.2.0 or higher installed. This also means that using the <code>StableCascadeCombinedPipeline</code> with <code>torch.bfloat16</code> requires PyTorch 2.2.0 or higher, since it calls the <code>StableCascadeDecoderPipeline</code> internally.</p> <p>If it is not possible to install PyTorch 2.2.0 or higher in your environment, the <code>StableCascadeDecoderPipeline</code> can be used on its own with the <code>torch.float16</code> data type. You can download the full precision or <code>bf16</code> variant weights for the pipeline and cast the weights to <code>torch.float16</code>.</p>",nt,K,at,ee,ot,te,it,ne,st,ae,rt,oe,dn="Loading the original format checkpoints is supported via <code>from_single_file</code> method in the StableCascadeUNet.",lt,ie,dt,se,ct,re,pt,le,cn="The model is intended for research purposes for now. Possible research areas and tasks include",mt,de,pn="<li>Research on generative models.</li> <li>Safe deployment of models which have the potential to generate harmful content.</li> <li>Probing and understanding the limitations and biases of generative models.</li> <li>Generation of artworks and use in design and other artistic processes.</li> <li>Applications in educational or creative tools.</li>",gt,ce,mn="Excluded uses are described below.",ut,pe,ft,me,gn=`The model was not trained to be factual or true representations of people or events,
and therefore using the model to generate such content is out-of-scope for the abilities of this model.
The model should not be used in any way that violates Stability AI’s <a href="https://stability.ai/use-policy" rel="nofollow">Acceptable Use Policy</a>.`,_t,ge,ht,ue,bt,fe,un="<li>Faces and people in general may not be generated properly.</li> <li>The autoencoding part of the model is lossy.</li>",yt,_e,wt,T,he,Zt,Se,fn="Combined Pipeline for text-to-image generation using Stable Cascade.",$t,Ze,_n=`This model inherits from <a href="/docs/diffusers/pr_12595/en/api/pipelines/overview#diffusers.DiffusionPipeline">DiffusionPipeline</a>. Check the superclass documentation for the generic methods the
library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)`,kt,j,be,Wt,$e,hn="Function invoked when calling the pipeline for generation.",Gt,N,Bt,V,ye,Nt,ke,bn=`Offloads all models to CPU using accelerate, reducing memory usage with a low impact on performance. Compared
to <code>enable_sequential_cpu_offload</code>, this method moves one whole model at a time to the GPU when its <code>forward</code>
method is called, and the model remains in GPU until the next model runs. Memory savings are lower than with
<code>enable_sequential_cpu_offload</code>, but performance is much better due to the iterative execution of the <code>unet</code>.`,Vt,Y,we,Yt,We,yn=`Offloads all models (<code>unet</code>, <code>text_encoder</code>, <code>vae</code>, and <code>safety checker</code> state dicts) to CPU using 🤗
Accelerate, significantly reducing memory usage. Models are moved to a <code>torch.device(&#39;meta&#39;)</code> and loaded on a
GPU only when their specific submodule’s <code>forward</code> method is called. Offloading happens on a submodule basis.
Memory savings are higher than using <code>enable_model_cpu_offload</code>, but performance is lower.`,Ct,Ce,Tt,U,Te,Rt,Ge,wn="Pipeline for generating image prior for Stable Cascade.",Lt,Be,Cn=`This model inherits from <a href="/docs/diffusers/pr_12595/en/api/pipelines/overview#diffusers.DiffusionPipeline">DiffusionPipeline</a>. Check the superclass documentation for the generic methods the
library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)`,Dt,S,Me,Et,Ne,Tn="Function invoked when calling the pipeline for generation.",Ht,R,Mt,ve,vt,$,Je,Qt,Ve,Mn="Output class for WuerstchenPriorPipeline.",Jt,Ue,Ut,P,Pe,Xt,Ye,vn="Pipeline for generating images from the Stable Cascade model.",zt,Re,Jn=`This model inherits from <a href="/docs/diffusers/pr_12595/en/api/pipelines/overview#diffusers.DiffusionPipeline">DiffusionPipeline</a>. Check the superclass documentation for the generic methods the
library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)`,Ft,Z,xe,qt,Le,Un="Function invoked when calling the pipeline for generation.",At,L,Pt,Ie,xt,Qe,It;return y=new Zn({props:{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"}}),C=new v({props:{title:"Stable Cascade",local:"stable-cascade",headingTag:"h1"}}),z=new v({props:{title:"Model Overview",local:"model-overview",headingTag:"h2"}}),K=new v({props:{title:"Usage example",local:"usage-example",headingTag:"h2"}}),ee=new He({props:{code:"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",highlighted:`<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableCascadeDecoderPipeline, StableCascadePriorPipeline

prompt = <span class="hljs-string">&quot;an image of a shiba inu, donning a spacesuit and helmet&quot;</span>
negative_prompt = <span class="hljs-string">&quot;&quot;</span>

prior = StableCascadePriorPipeline.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade-prior&quot;</span>, variant=<span class="hljs-string">&quot;bf16&quot;</span>, torch_dtype=torch.bfloat16)
decoder = StableCascadeDecoderPipeline.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade&quot;</span>, variant=<span class="hljs-string">&quot;bf16&quot;</span>, torch_dtype=torch.float16)

prior.enable_model_cpu_offload()
prior_output = prior(
    prompt=prompt,
    height=<span class="hljs-number">1024</span>,
    width=<span class="hljs-number">1024</span>,
    negative_prompt=negative_prompt,
    guidance_scale=<span class="hljs-number">4.0</span>,
    num_images_per_prompt=<span class="hljs-number">1</span>,
    num_inference_steps=<span class="hljs-number">20</span>
)

decoder.enable_model_cpu_offload()
decoder_output = decoder(
    image_embeddings=prior_output.image_embeddings.to(torch.float16),
    prompt=prompt,
    negative_prompt=negative_prompt,
    guidance_scale=<span class="hljs-number">0.0</span>,
    output_type=<span class="hljs-string">&quot;pil&quot;</span>,
    num_inference_steps=<span class="hljs-number">10</span>
).images[<span class="hljs-number">0</span>]
decoder_output.save(<span class="hljs-string">&quot;cascade.png&quot;</span>)`,wrap:!1}}),te=new v({props:{title:"Using the Lite Versions of the Stage B and Stage C models",local:"using-the-lite-versions-of-the-stage-b-and-stage-c-models",headingTag:"h2"}}),ne=new He({props:{code:"aW1wb3J0JTIwdG9yY2glMEFmcm9tJTIwZGlmZnVzZXJzJTIwaW1wb3J0JTIwKCUwQSUyMCUyMCUyMCUyMFN0YWJsZUNhc2NhZGVEZWNvZGVyUGlwZWxpbmUlMkMlMEElMjAlMjAlMjAlMjBTdGFibGVDYXNjYWRlUHJpb3JQaXBlbGluZSUyQyUwQSUyMCUyMCUyMCUyMFN0YWJsZUNhc2NhZGVVTmV0JTJDJTBBKSUwQSUwQXByb21wdCUyMCUzRCUyMCUyMmFuJTIwaW1hZ2UlMjBvZiUyMGElMjBzaGliYSUyMGludSUyQyUyMGRvbm5pbmclMjBhJTIwc3BhY2VzdWl0JTIwYW5kJTIwaGVsbWV0JTIyJTBBbmVnYXRpdmVfcHJvbXB0JTIwJTNEJTIwJTIyJTIyJTBBJTBBcHJpb3JfdW5ldCUyMCUzRCUyMFN0YWJsZUNhc2NhZGVVTmV0LmZyb21fcHJldHJhaW5lZCglMjJzdGFiaWxpdHlhaSUyRnN0YWJsZS1jYXNjYWRlLXByaW9yJTIyJTJDJTIwc3ViZm9sZGVyJTNEJTIycHJpb3JfbGl0ZSUyMiklMEFkZWNvZGVyX3VuZXQlMjAlM0QlMjBTdGFibGVDYXNjYWRlVU5ldC5mcm9tX3ByZXRyYWluZWQoJTIyc3RhYmlsaXR5YWklMkZzdGFibGUtY2FzY2FkZSUyMiUyQyUyMHN1YmZvbGRlciUzRCUyMmRlY29kZXJfbGl0ZSUyMiklMEElMEFwcmlvciUyMCUzRCUyMFN0YWJsZUNhc2NhZGVQcmlvclBpcGVsaW5lLmZyb21fcHJldHJhaW5lZCglMjJzdGFiaWxpdHlhaSUyRnN0YWJsZS1jYXNjYWRlLXByaW9yJTIyJTJDJTIwcHJpb3IlM0Rwcmlvcl91bmV0KSUwQWRlY29kZXIlMjAlM0QlMjBTdGFibGVDYXNjYWRlRGVjb2RlclBpcGVsaW5lLmZyb21fcHJldHJhaW5lZCglMjJzdGFiaWxpdHlhaSUyRnN0YWJsZS1jYXNjYWRlJTIyJTJDJTIwZGVjb2RlciUzRGRlY29kZXJfdW5ldCklMEElMEFwcmlvci5lbmFibGVfbW9kZWxfY3B1X29mZmxvYWQoKSUwQXByaW9yX291dHB1dCUyMCUzRCUyMHByaW9yKCUwQSUyMCUyMCUyMCUyMHByb21wdCUzRHByb21wdCUyQyUwQSUyMCUyMCUyMCUyMGhlaWdodCUzRDEwMjQlMkMlMEElMjAlMjAlMjAlMjB3aWR0aCUzRDEwMjQlMkMlMEElMjAlMjAlMjAlMjBuZWdhdGl2ZV9wcm9tcHQlM0RuZWdhdGl2ZV9wcm9tcHQlMkMlMEElMjAlMjAlMjAlMjBndWlkYW5jZV9zY2FsZSUzRDQuMCUyQyUwQSUyMCUyMCUyMCUyMG51bV9pbWFnZXNfcGVyX3Byb21wdCUzRDElMkMlMEElMjAlMjAlMjAlMjBudW1faW5mZXJlbmNlX3N0ZXBzJTNEMjAlMEEpJTBBJTBBZGVjb2Rlci5lbmFibGVfbW9kZWxfY3B1X29mZmxvYWQoKSUwQWRlY29kZXJfb3V0cHV0JTIwJTNEJTIwZGVjb2RlciglMEElMjAlMjAlMjAlMjBpbWFnZV9lbWJlZGRpbmdzJTNEcHJpb3Jfb3V0cHV0LmltYWdlX2VtYmVkZGluZ3MlMkMlMEElMjAlMjAlMjAlMjBwcm9tcHQlM0Rwcm9tcHQlMkMlMEElMjAlMjAlMjAlMjBuZWdhdGl2ZV9wcm9tcHQlM0RuZWdhdGl2ZV9wcm9tcHQlMkMlMEElMjAlMjAlMjAlMjBndWlkYW5jZV9zY2FsZSUzRDAuMCUyQyUwQSUyMCUyMCUyMCUyMG91dHB1dF90eXBlJTNEJTIycGlsJTIyJTJDJTBBJTIwJTIwJTIwJTIwbnVtX2luZmVyZW5jZV9zdGVwcyUzRDEwJTBBKS5pbWFnZXMlNUIwJTVEJTBBZGVjb2Rlcl9vdXRwdXQuc2F2ZSglMjJjYXNjYWRlLnBuZyUyMik=",highlighted:`<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> (
    StableCascadeDecoderPipeline,
    StableCascadePriorPipeline,
    StableCascadeUNet,
)

prompt = <span class="hljs-string">&quot;an image of a shiba inu, donning a spacesuit and helmet&quot;</span>
negative_prompt = <span class="hljs-string">&quot;&quot;</span>

prior_unet = StableCascadeUNet.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade-prior&quot;</span>, subfolder=<span class="hljs-string">&quot;prior_lite&quot;</span>)
decoder_unet = StableCascadeUNet.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade&quot;</span>, subfolder=<span class="hljs-string">&quot;decoder_lite&quot;</span>)

prior = StableCascadePriorPipeline.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade-prior&quot;</span>, prior=prior_unet)
decoder = StableCascadeDecoderPipeline.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade&quot;</span>, decoder=decoder_unet)

prior.enable_model_cpu_offload()
prior_output = prior(
    prompt=prompt,
    height=<span class="hljs-number">1024</span>,
    width=<span class="hljs-number">1024</span>,
    negative_prompt=negative_prompt,
    guidance_scale=<span class="hljs-number">4.0</span>,
    num_images_per_prompt=<span class="hljs-number">1</span>,
    num_inference_steps=<span class="hljs-number">20</span>
)

decoder.enable_model_cpu_offload()
decoder_output = decoder(
    image_embeddings=prior_output.image_embeddings,
    prompt=prompt,
    negative_prompt=negative_prompt,
    guidance_scale=<span class="hljs-number">0.0</span>,
    output_type=<span class="hljs-string">&quot;pil&quot;</span>,
    num_inference_steps=<span class="hljs-number">10</span>
).images[<span class="hljs-number">0</span>]
decoder_output.save(<span class="hljs-string">&quot;cascade.png&quot;</span>)`,wrap:!1}}),ae=new v({props:{title:"Loading original checkpoints with from_single_file",local:"loading-original-checkpoints-with-fromsinglefile",headingTag:"h2"}}),ie=new He({props:{code:"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",highlighted:`<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> (
    StableCascadeDecoderPipeline,
    StableCascadePriorPipeline,
    StableCascadeUNet,
)

prompt = <span class="hljs-string">&quot;an image of a shiba inu, donning a spacesuit and helmet&quot;</span>
negative_prompt = <span class="hljs-string">&quot;&quot;</span>

prior_unet = StableCascadeUNet.from_single_file(
    <span class="hljs-string">&quot;https://huggingface.co/stabilityai/stable-cascade/resolve/main/stage_c_bf16.safetensors&quot;</span>,
    torch_dtype=torch.bfloat16
)
decoder_unet = StableCascadeUNet.from_single_file(
    <span class="hljs-string">&quot;https://huggingface.co/stabilityai/stable-cascade/blob/main/stage_b_bf16.safetensors&quot;</span>,
    torch_dtype=torch.bfloat16
)

prior = StableCascadePriorPipeline.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade-prior&quot;</span>, prior=prior_unet, torch_dtype=torch.bfloat16)
decoder = StableCascadeDecoderPipeline.from_pretrained(<span class="hljs-string">&quot;stabilityai/stable-cascade&quot;</span>, decoder=decoder_unet, torch_dtype=torch.bfloat16)

prior.enable_model_cpu_offload()
prior_output = prior(
    prompt=prompt,
    height=<span class="hljs-number">1024</span>,
    width=<span class="hljs-number">1024</span>,
    negative_prompt=negative_prompt,
    guidance_scale=<span class="hljs-number">4.0</span>,
    num_images_per_prompt=<span class="hljs-number">1</span>,
    num_inference_steps=<span class="hljs-number">20</span>
)

decoder.enable_model_cpu_offload()
decoder_output = decoder(
    image_embeddings=prior_output.image_embeddings,
    prompt=prompt,
    negative_prompt=negative_prompt,
    guidance_scale=<span class="hljs-number">0.0</span>,
    output_type=<span class="hljs-string">&quot;pil&quot;</span>,
    num_inference_steps=<span class="hljs-number">10</span>
).images[<span class="hljs-number">0</span>]
decoder_output.save(<span class="hljs-string">&quot;cascade-single-file.png&quot;</span>)`,wrap:!1}}),se=new v({props:{title:"Uses",local:"uses",headingTag:"h2"}}),re=new v({props:{title:"Direct Use",local:"direct-use",headingTag:"h3"}}),pe=new v({props:{title:"Out-of-Scope Use",local:"out-of-scope-use",headingTag:"h3"}}),ge=new v({props:{title:"Limitations and Bias",local:"limitations-and-bias",headingTag:"h2"}}),ue=new v({props:{title:"Limitations",local:"limitations",headingTag:"h3"}}),_e=new v({props:{title:"StableCascadeCombinedPipeline",local:"diffusers.StableCascadeCombinedPipeline",headingTag:"h2"}}),he=new W({props:{name:"class diffusers.StableCascadeCombinedPipeline",anchor:"diffusers.StableCascadeCombinedPipeline",parameters:[{name:"tokenizer",val:": CLIPTokenizer"},{name:"text_encoder",val:": CLIPTextModelWithProjection"},{name:"decoder",val:": StableCascadeUNet"},{name:"scheduler",val:": DDPMWuerstchenScheduler"},{name:"vqgan",val:": PaellaVQModel"},{name:"prior_prior",val:": StableCascadeUNet"},{name:"prior_text_encoder",val:": CLIPTextModelWithProjection"},{name:"prior_tokenizer",val:": CLIPTokenizer"},{name:"prior_scheduler",val:": DDPMWuerstchenScheduler"},{name:"prior_feature_extractor",val:": typing.Optional[transformers.models.clip.image_processing_clip.CLIPImageProcessor] = None"},{name:"prior_image_encoder",val:": typing.Optional[transformers.models.clip.modeling_clip.CLIPVisionModelWithProjection] = None"}],parametersDescription:[{anchor:"diffusers.StableCascadeCombinedPipeline.tokenizer",description:`<strong>tokenizer</strong> (<code>CLIPTokenizer</code>) &#x2014;
The decoder tokenizer to be used for text inputs.`,name:"tokenizer"},{anchor:"diffusers.StableCascadeCombinedPipeline.text_encoder",description:`<strong>text_encoder</strong> (<code>CLIPTextModelWithProjection</code>) &#x2014;
The decoder text encoder to be used for text inputs.`,name:"text_encoder"},{anchor:"diffusers.StableCascadeCombinedPipeline.decoder",description:`<strong>decoder</strong> (<code>StableCascadeUNet</code>) &#x2014;
The decoder model to be used for decoder image generation pipeline.`,name:"decoder"},{anchor:"diffusers.StableCascadeCombinedPipeline.scheduler",description:`<strong>scheduler</strong> (<code>DDPMWuerstchenScheduler</code>) &#x2014;
The scheduler to be used for decoder image generation pipeline.`,name:"scheduler"},{anchor:"diffusers.StableCascadeCombinedPipeline.vqgan",description:`<strong>vqgan</strong> (<code>PaellaVQModel</code>) &#x2014;
The VQGAN model to be used for decoder image generation pipeline.`,name:"vqgan"},{anchor:"diffusers.StableCascadeCombinedPipeline.prior_prior",description:`<strong>prior_prior</strong> (<code>StableCascadeUNet</code>) &#x2014;
The prior model to be used for prior pipeline.`,name:"prior_prior"},{anchor:"diffusers.StableCascadeCombinedPipeline.prior_text_encoder",description:`<strong>prior_text_encoder</strong> (<code>CLIPTextModelWithProjection</code>) &#x2014;
The prior text encoder to be used for text inputs.`,name:"prior_text_encoder"},{anchor:"diffusers.StableCascadeCombinedPipeline.prior_tokenizer",description:`<strong>prior_tokenizer</strong> (<code>CLIPTokenizer</code>) &#x2014;
The prior tokenizer to be used for text inputs.`,name:"prior_tokenizer"},{anchor:"diffusers.StableCascadeCombinedPipeline.prior_scheduler",description:`<strong>prior_scheduler</strong> (<code>DDPMWuerstchenScheduler</code>) &#x2014;
The scheduler to be used for prior pipeline.`,name:"prior_scheduler"},{anchor:"diffusers.StableCascadeCombinedPipeline.prior_feature_extractor",description:`<strong>prior_feature_extractor</strong> (<a href="https://huggingface.co/docs/transformers/main/en/model_doc/clip#transformers.CLIPImageProcessor" rel="nofollow">CLIPImageProcessor</a>) &#x2014;
Model that extracts features from generated images to be used as inputs for the <code>image_encoder</code>.`,name:"prior_feature_extractor"},{anchor:"diffusers.StableCascadeCombinedPipeline.prior_image_encoder",description:`<strong>prior_image_encoder</strong> (<code>CLIPVisionModelWithProjection</code>) &#x2014;
Frozen CLIP image-encoder (<a href="https://huggingface.co/openai/clip-vit-large-patch14" rel="nofollow">clip-vit-large-patch14</a>).`,name:"prior_image_encoder"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_combined.py#L45"}}),be=new W({props:{name:"__call__",anchor:"diffusers.StableCascadeCombinedPipeline.__call__",parameters:[{name:"prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"images",val:": typing.Union[torch.Tensor, PIL.Image.Image, typing.List[torch.Tensor], typing.List[PIL.Image.Image]] = None"},{name:"height",val:": int = 512"},{name:"width",val:": int = 512"},{name:"prior_num_inference_steps",val:": int = 60"},{name:"prior_guidance_scale",val:": float = 4.0"},{name:"num_inference_steps",val:": int = 12"},{name:"decoder_guidance_scale",val:": float = 0.0"},{name:"negative_prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"prompt_embeds_pooled",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds_pooled",val:": typing.Optional[torch.Tensor] = None"},{name:"num_images_per_prompt",val:": int = 1"},{name:"generator",val:": typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None"},{name:"latents",val:": typing.Optional[torch.Tensor] = None"},{name:"output_type",val:": typing.Optional[str] = 'pil'"},{name:"return_dict",val:": bool = True"},{name:"prior_callback_on_step_end",val:": typing.Optional[typing.Callable[[int, int, typing.Dict], NoneType]] = None"},{name:"prior_callback_on_step_end_tensor_inputs",val:": typing.List[str] = ['latents']"},{name:"callback_on_step_end",val:": typing.Optional[typing.Callable[[int, int, typing.Dict], NoneType]] = None"},{name:"callback_on_step_end_tensor_inputs",val:": typing.List[str] = ['latents']"}],parametersDescription:[{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>) &#x2014;
The prompt or prompts to guide the image generation for the prior and decoder.`,name:"prompt"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.images",description:`<strong>images</strong> (<code>torch.Tensor</code>, <code>PIL.Image.Image</code>, <code>List[torch.Tensor]</code>, <code>List[PIL.Image.Image]</code>, <em>optional</em>) &#x2014;
The images to guide the image generation for the prior.`,name:"images"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.negative_prompt",description:`<strong>negative_prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
if <code>guidance_scale</code> is less than <code>1</code>).`,name:"negative_prompt"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prompt_embeds",description:`<strong>prompt_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated text embeddings for the prior. Can be used to easily tweak text inputs, <em>e.g.</em> prompt
weighting. If not provided, text embeddings will be generated from <code>prompt</code> input argument.`,name:"prompt_embeds"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prompt_embeds_pooled",description:`<strong>prompt_embeds_pooled</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated text embeddings for the prior. Can be used to easily tweak text inputs, <em>e.g.</em> prompt
weighting. If not provided, text embeddings will be generated from <code>prompt</code> input argument.`,name:"prompt_embeds_pooled"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.negative_prompt_embeds",description:`<strong>negative_prompt_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated negative text embeddings for the prior. Can be used to easily tweak text inputs, <em>e.g.</em>
prompt weighting. If not provided, negative_prompt_embeds will be generated from <code>negative_prompt</code>
input argument.`,name:"negative_prompt_embeds"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.negative_prompt_embeds_pooled",description:`<strong>negative_prompt_embeds_pooled</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated negative text embeddings for the prior. Can be used to easily tweak text inputs, <em>e.g.</em>
prompt weighting. If not provided, negative_prompt_embeds will be generated from <code>negative_prompt</code>
input argument.`,name:"negative_prompt_embeds_pooled"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.num_images_per_prompt",description:`<strong>num_images_per_prompt</strong> (<code>int</code>, <em>optional</em>, defaults to 1) &#x2014;
The number of images to generate per prompt.`,name:"num_images_per_prompt"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.height",description:`<strong>height</strong> (<code>int</code>, <em>optional</em>, defaults to 512) &#x2014;
The height in pixels of the generated image.`,name:"height"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.width",description:`<strong>width</strong> (<code>int</code>, <em>optional</em>, defaults to 512) &#x2014;
The width in pixels of the generated image.`,name:"width"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prior_guidance_scale",description:`<strong>prior_guidance_scale</strong> (<code>float</code>, <em>optional</em>, defaults to 4.0) &#x2014;
Guidance scale as defined in <a href="https://huggingface.co/papers/2207.12598" rel="nofollow">Classifier-Free Diffusion
Guidance</a>. <code>prior_guidance_scale</code> is defined as <code>w</code> of
equation 2. of <a href="https://huggingface.co/papers/2205.11487" rel="nofollow">Imagen Paper</a>. Guidance scale is enabled by
setting <code>prior_guidance_scale &gt; 1</code>. Higher guidance scale encourages to generate images that are
closely linked to the text <code>prompt</code>, usually at the expense of lower image quality.`,name:"prior_guidance_scale"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prior_num_inference_steps",description:`<strong>prior_num_inference_steps</strong> (<code>Union[int, Dict[float, int]]</code>, <em>optional</em>, defaults to 60) &#x2014;
The number of prior denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference. For more specific timestep spacing, you can pass customized
<code>prior_timesteps</code>`,name:"prior_num_inference_steps"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.num_inference_steps",description:`<strong>num_inference_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 12) &#x2014;
The number of decoder denoising steps. More denoising steps usually lead to a higher quality image at
the expense of slower inference. For more specific timestep spacing, you can pass customized
<code>timesteps</code>`,name:"num_inference_steps"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.decoder_guidance_scale",description:`<strong>decoder_guidance_scale</strong> (<code>float</code>, <em>optional</em>, defaults to 0.0) &#x2014;
Guidance scale as defined in <a href="https://huggingface.co/papers/2207.12598" rel="nofollow">Classifier-Free Diffusion
Guidance</a>. <code>guidance_scale</code> is defined as <code>w</code> of equation 2.
of <a href="https://huggingface.co/papers/2205.11487" rel="nofollow">Imagen Paper</a>. Guidance scale is enabled by setting
<code>guidance_scale &gt; 1</code>. Higher guidance scale encourages to generate images that are closely linked to
the text <code>prompt</code>, usually at the expense of lower image quality.`,name:"decoder_guidance_scale"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.generator",description:`<strong>generator</strong> (<code>torch.Generator</code> or <code>List[torch.Generator]</code>, <em>optional</em>) &#x2014;
One or a list of <a href="https://pytorch.org/docs/stable/generated/torch.Generator.html" rel="nofollow">torch generator(s)</a>
to make generation deterministic.`,name:"generator"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.latents",description:`<strong>latents</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will be generated by sampling using the supplied random <code>generator</code>.`,name:"latents"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.output_type",description:`<strong>output_type</strong> (<code>str</code>, <em>optional</em>, defaults to <code>&quot;pil&quot;</code>) &#x2014;
The output format of the generate image. Choose between: <code>&quot;pil&quot;</code> (<code>PIL.Image.Image</code>), <code>&quot;np&quot;</code>
(<code>np.array</code>) or <code>&quot;pt&quot;</code> (<code>torch.Tensor</code>).`,name:"output_type"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether or not to return a <a href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput">ImagePipelineOutput</a> instead of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prior_callback_on_step_end",description:`<strong>prior_callback_on_step_end</strong> (<code>Callable</code>, <em>optional</em>) &#x2014;
A function that calls at the end of each denoising steps during the inference. The function is called
with the following arguments: <code>prior_callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)</code>.`,name:"prior_callback_on_step_end"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.prior_callback_on_step_end_tensor_inputs",description:`<strong>prior_callback_on_step_end_tensor_inputs</strong> (<code>List</code>, <em>optional</em>) &#x2014;
The list of tensor inputs for the <code>prior_callback_on_step_end</code> function. The tensors specified in the
list will be passed as <code>callback_kwargs</code> argument. You will only be able to include variables listed in
the <code>._callback_tensor_inputs</code> attribute of your pipeline class.`,name:"prior_callback_on_step_end_tensor_inputs"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.callback_on_step_end",description:`<strong>callback_on_step_end</strong> (<code>Callable</code>, <em>optional</em>) &#x2014;
A function that calls at the end of each denoising steps during the inference. The function is called
with the following arguments: <code>callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)</code>. <code>callback_kwargs</code> will include a list of all tensors as specified by
<code>callback_on_step_end_tensor_inputs</code>.`,name:"callback_on_step_end"},{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.callback_on_step_end_tensor_inputs",description:`<strong>callback_on_step_end_tensor_inputs</strong> (<code>List</code>, <em>optional</em>) &#x2014;
The list of tensor inputs for the <code>callback_on_step_end</code> function. The tensors specified in the list
will be passed as <code>callback_kwargs</code> argument. You will only be able to include variables listed in the
<code>._callback_tensor_inputs</code> attribute of your pipeline class.`,name:"callback_on_step_end_tensor_inputs"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_combined.py#L158",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


<p><a
  href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput"
>ImagePipelineOutput</a> or <code>tuple</code> <a
  href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput"
>ImagePipelineOutput</a> if <code>return_dict</code> is True,
otherwise a <code>tuple</code>. When returning a tuple, the first element is a list with the generated images.</p>
`}}),N=new Ot({props:{anchor:"diffusers.StableCascadeCombinedPipeline.__call__.example",$$slots:{default:[kn]},$$scope:{ctx:G}}}),ye=new W({props:{name:"enable_model_cpu_offload",anchor:"diffusers.StableCascadeCombinedPipeline.enable_model_cpu_offload",parameters:[{name:"gpu_id",val:": typing.Optional[int] = None"},{name:"device",val:": typing.Union[torch.device, str] = None"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_combined.py#L130"}}),we=new W({props:{name:"enable_sequential_cpu_offload",anchor:"diffusers.StableCascadeCombinedPipeline.enable_sequential_cpu_offload",parameters:[{name:"gpu_id",val:": typing.Optional[int] = None"},{name:"device",val:": typing.Union[torch.device, str] = None"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_combined.py#L140"}}),Ce=new v({props:{title:"StableCascadePriorPipeline",local:"diffusers.StableCascadePriorPipeline",headingTag:"h2"}}),Te=new W({props:{name:"class diffusers.StableCascadePriorPipeline",anchor:"diffusers.StableCascadePriorPipeline",parameters:[{name:"tokenizer",val:": CLIPTokenizer"},{name:"text_encoder",val:": CLIPTextModelWithProjection"},{name:"prior",val:": StableCascadeUNet"},{name:"scheduler",val:": DDPMWuerstchenScheduler"},{name:"resolution_multiple",val:": float = 42.67"},{name:"feature_extractor",val:": typing.Optional[transformers.models.clip.image_processing_clip.CLIPImageProcessor] = None"},{name:"image_encoder",val:": typing.Optional[transformers.models.clip.modeling_clip.CLIPVisionModelWithProjection] = None"}],parametersDescription:[{anchor:"diffusers.StableCascadePriorPipeline.prior",description:`<strong>prior</strong> (<code>StableCascadeUNet</code>) &#x2014;
The Stable Cascade prior to approximate the image embedding from the text and/or image embedding.`,name:"prior"},{anchor:"diffusers.StableCascadePriorPipeline.text_encoder",description:`<strong>text_encoder</strong> (<code>CLIPTextModelWithProjection</code>) &#x2014;
Frozen text-encoder
(<a href="https://huggingface.co/laion/CLIP-ViT-bigG-14-laion2B-39B-b160k" rel="nofollow">laion/CLIP-ViT-bigG-14-laion2B-39B-b160k</a>).`,name:"text_encoder"},{anchor:"diffusers.StableCascadePriorPipeline.feature_extractor",description:`<strong>feature_extractor</strong> (<a href="https://huggingface.co/docs/transformers/main/en/model_doc/clip#transformers.CLIPImageProcessor" rel="nofollow">CLIPImageProcessor</a>) &#x2014;
Model that extracts features from generated images to be used as inputs for the <code>image_encoder</code>.`,name:"feature_extractor"},{anchor:"diffusers.StableCascadePriorPipeline.image_encoder",description:`<strong>image_encoder</strong> (<code>CLIPVisionModelWithProjection</code>) &#x2014;
Frozen CLIP image-encoder (<a href="https://huggingface.co/openai/clip-vit-large-patch14" rel="nofollow">clip-vit-large-patch14</a>).`,name:"image_encoder"},{anchor:"diffusers.StableCascadePriorPipeline.tokenizer",description:`<strong>tokenizer</strong> (<code>CLIPTokenizer</code>) &#x2014;
Tokenizer of class
<a href="https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer" rel="nofollow">CLIPTokenizer</a>.`,name:"tokenizer"},{anchor:"diffusers.StableCascadePriorPipeline.scheduler",description:`<strong>scheduler</strong> (<code>DDPMWuerstchenScheduler</code>) &#x2014;
A scheduler to be used in combination with <code>prior</code> to generate image embedding.`,name:"scheduler"},{anchor:"diffusers.StableCascadePriorPipeline.resolution_multiple",description:`<strong>resolution_multiple</strong> (&#x2018;float&#x2019;, <em>optional</em>, defaults to 42.67) &#x2014;
Default resolution for multiple images generated.`,name:"resolution_multiple"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_prior.py#L80"}}),Me=new W({props:{name:"__call__",anchor:"diffusers.StableCascadePriorPipeline.__call__",parameters:[{name:"prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"images",val:": typing.Union[torch.Tensor, PIL.Image.Image, typing.List[torch.Tensor], typing.List[PIL.Image.Image]] = None"},{name:"height",val:": int = 1024"},{name:"width",val:": int = 1024"},{name:"num_inference_steps",val:": int = 20"},{name:"timesteps",val:": typing.List[float] = None"},{name:"guidance_scale",val:": float = 4.0"},{name:"negative_prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"prompt_embeds_pooled",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds_pooled",val:": typing.Optional[torch.Tensor] = None"},{name:"image_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"num_images_per_prompt",val:": typing.Optional[int] = 1"},{name:"generator",val:": typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None"},{name:"latents",val:": typing.Optional[torch.Tensor] = None"},{name:"output_type",val:": typing.Optional[str] = 'pt'"},{name:"return_dict",val:": bool = True"},{name:"callback_on_step_end",val:": typing.Optional[typing.Callable[[int, int, typing.Dict], NoneType]] = None"},{name:"callback_on_step_end_tensor_inputs",val:": typing.List[str] = ['latents']"}],parametersDescription:[{anchor:"diffusers.StableCascadePriorPipeline.__call__.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>) &#x2014;
The prompt or prompts to guide the image generation.`,name:"prompt"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.height",description:`<strong>height</strong> (<code>int</code>, <em>optional</em>, defaults to 1024) &#x2014;
The height in pixels of the generated image.`,name:"height"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.width",description:`<strong>width</strong> (<code>int</code>, <em>optional</em>, defaults to 1024) &#x2014;
The width in pixels of the generated image.`,name:"width"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.num_inference_steps",description:`<strong>num_inference_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 60) &#x2014;
The number of denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference.`,name:"num_inference_steps"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.guidance_scale",description:`<strong>guidance_scale</strong> (<code>float</code>, <em>optional</em>, defaults to 8.0) &#x2014;
Guidance scale as defined in <a href="https://huggingface.co/papers/2207.12598" rel="nofollow">Classifier-Free Diffusion
Guidance</a>. <code>decoder_guidance_scale</code> is defined as <code>w</code> of
equation 2. of <a href="https://huggingface.co/papers/2205.11487" rel="nofollow">Imagen Paper</a>. Guidance scale is enabled by
setting <code>decoder_guidance_scale &gt; 1</code>. Higher guidance scale encourages to generate images that are
closely linked to the text <code>prompt</code>, usually at the expense of lower image quality.`,name:"guidance_scale"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.negative_prompt",description:`<strong>negative_prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
if <code>decoder_guidance_scale</code> is less than <code>1</code>).`,name:"negative_prompt"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.prompt_embeds",description:`<strong>prompt_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt weighting. If not
provided, text embeddings will be generated from <code>prompt</code> input argument.`,name:"prompt_embeds"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.prompt_embeds_pooled",description:`<strong>prompt_embeds_pooled</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt weighting.
If not provided, pooled text embeddings will be generated from <code>prompt</code> input argument.`,name:"prompt_embeds_pooled"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.negative_prompt_embeds",description:`<strong>negative_prompt_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt
weighting. If not provided, negative_prompt_embeds will be generated from <code>negative_prompt</code> input
argument.`,name:"negative_prompt_embeds"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.negative_prompt_embeds_pooled",description:`<strong>negative_prompt_embeds_pooled</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt
weighting. If not provided, negative_prompt_embeds_pooled will be generated from <code>negative_prompt</code>
input argument.`,name:"negative_prompt_embeds_pooled"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.image_embeds",description:`<strong>image_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated image embeddings. Can be used to easily tweak image inputs, <em>e.g.</em> prompt weighting. If
not provided, image embeddings will be generated from <code>image</code> input argument if existing.`,name:"image_embeds"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.num_images_per_prompt",description:`<strong>num_images_per_prompt</strong> (<code>int</code>, <em>optional</em>, defaults to 1) &#x2014;
The number of images to generate per prompt.`,name:"num_images_per_prompt"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.generator",description:`<strong>generator</strong> (<code>torch.Generator</code> or <code>List[torch.Generator]</code>, <em>optional</em>) &#x2014;
One or a list of <a href="https://pytorch.org/docs/stable/generated/torch.Generator.html" rel="nofollow">torch generator(s)</a>
to make generation deterministic.`,name:"generator"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.latents",description:`<strong>latents</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will be generated by sampling using the supplied random <code>generator</code>.`,name:"latents"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.output_type",description:`<strong>output_type</strong> (<code>str</code>, <em>optional</em>, defaults to <code>&quot;pil&quot;</code>) &#x2014;
The output format of the generate image. Choose between: <code>&quot;pil&quot;</code> (<code>PIL.Image.Image</code>), <code>&quot;np&quot;</code>
(<code>np.array</code>) or <code>&quot;pt&quot;</code> (<code>torch.Tensor</code>).`,name:"output_type"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether or not to return a <a href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput">ImagePipelineOutput</a> instead of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.callback_on_step_end",description:`<strong>callback_on_step_end</strong> (<code>Callable</code>, <em>optional</em>) &#x2014;
A function that calls at the end of each denoising steps during the inference. The function is called
with the following arguments: <code>callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)</code>. <code>callback_kwargs</code> will include a list of all tensors as specified by
<code>callback_on_step_end_tensor_inputs</code>.`,name:"callback_on_step_end"},{anchor:"diffusers.StableCascadePriorPipeline.__call__.callback_on_step_end_tensor_inputs",description:`<strong>callback_on_step_end_tensor_inputs</strong> (<code>List</code>, <em>optional</em>) &#x2014;
The list of tensor inputs for the <code>callback_on_step_end</code> function. The tensors specified in the list
will be passed as <code>callback_kwargs</code> argument. You will only be able to include variables listed in the
<code>._callback_tensor_inputs</code> attribute of your pipeline class.`,name:"callback_on_step_end_tensor_inputs"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_prior.py#L375",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


<p><code>StableCascadePriorPipelineOutput</code> or <code>tuple</code> <code>StableCascadePriorPipelineOutput</code> if <code>return_dict</code> is
True, otherwise a <code>tuple</code>. When returning a tuple, the first element is a list with the generated image
embeddings.</p>
`}}),R=new Ot({props:{anchor:"diffusers.StableCascadePriorPipeline.__call__.example",$$slots:{default:[Wn]},$$scope:{ctx:G}}}),ve=new v({props:{title:"StableCascadePriorPipelineOutput",local:"diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput",headingTag:"h2"}}),Je=new W({props:{name:"class diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput",anchor:"diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput",parameters:[{name:"image_embeddings",val:": typing.Union[torch.Tensor, numpy.ndarray]"},{name:"prompt_embeds",val:": typing.Union[torch.Tensor, numpy.ndarray]"},{name:"prompt_embeds_pooled",val:": typing.Union[torch.Tensor, numpy.ndarray]"},{name:"negative_prompt_embeds",val:": typing.Union[torch.Tensor, numpy.ndarray]"},{name:"negative_prompt_embeds_pooled",val:": typing.Union[torch.Tensor, numpy.ndarray]"}],parametersDescription:[{anchor:"diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput.image_embeddings",description:`<strong>image_embeddings</strong> (<code>torch.Tensor</code> or <code>np.ndarray</code>) &#x2014;
Prior image embeddings for text prompt`,name:"image_embeddings"},{anchor:"diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput.prompt_embeds",description:`<strong>prompt_embeds</strong> (<code>torch.Tensor</code>) &#x2014;
Text embeddings for the prompt.`,name:"prompt_embeds"},{anchor:"diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput.negative_prompt_embeds",description:`<strong>negative_prompt_embeds</strong> (<code>torch.Tensor</code>) &#x2014;
Text embeddings for the negative prompt.`,name:"negative_prompt_embeds"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade_prior.py#L60"}}),Ue=new v({props:{title:"StableCascadeDecoderPipeline",local:"diffusers.StableCascadeDecoderPipeline",headingTag:"h2"}}),Pe=new W({props:{name:"class diffusers.StableCascadeDecoderPipeline",anchor:"diffusers.StableCascadeDecoderPipeline",parameters:[{name:"decoder",val:": StableCascadeUNet"},{name:"tokenizer",val:": CLIPTokenizer"},{name:"text_encoder",val:": CLIPTextModelWithProjection"},{name:"scheduler",val:": DDPMWuerstchenScheduler"},{name:"vqgan",val:": PaellaVQModel"},{name:"latent_dim_scale",val:": float = 10.67"}],parametersDescription:[{anchor:"diffusers.StableCascadeDecoderPipeline.tokenizer",description:`<strong>tokenizer</strong> (<code>CLIPTokenizer</code>) &#x2014;
The CLIP tokenizer.`,name:"tokenizer"},{anchor:"diffusers.StableCascadeDecoderPipeline.text_encoder",description:`<strong>text_encoder</strong> (<code>CLIPTextModelWithProjection</code>) &#x2014;
The CLIP text encoder.`,name:"text_encoder"},{anchor:"diffusers.StableCascadeDecoderPipeline.decoder",description:`<strong>decoder</strong> (<code>StableCascadeUNet</code>) &#x2014;
The Stable Cascade decoder unet.`,name:"decoder"},{anchor:"diffusers.StableCascadeDecoderPipeline.vqgan",description:`<strong>vqgan</strong> (<code>PaellaVQModel</code>) &#x2014;
The VQGAN model.`,name:"vqgan"},{anchor:"diffusers.StableCascadeDecoderPipeline.scheduler",description:`<strong>scheduler</strong> (<code>DDPMWuerstchenScheduler</code>) &#x2014;
A scheduler to be used in combination with <code>prior</code> to generate image embedding.`,name:"scheduler"},{anchor:"diffusers.StableCascadeDecoderPipeline.latent_dim_scale",description:`<strong>latent_dim_scale</strong> (float, <code>optional</code>, defaults to 10.67) &#x2014;
Multiplier to determine the VQ latent space size from the image embeddings. If the image embeddings are
height=24 and width=24, the VQ latent shape needs to be height=int(24<em>10.67)=256 and
width=int(24</em>10.67)=256 in order to match the training conditions.`,name:"latent_dim_scale"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade.py#L58"}}),xe=new W({props:{name:"__call__",anchor:"diffusers.StableCascadeDecoderPipeline.__call__",parameters:[{name:"image_embeddings",val:": typing.Union[torch.Tensor, typing.List[torch.Tensor]]"},{name:"prompt",val:": typing.Union[str, typing.List[str]] = None"},{name:"num_inference_steps",val:": int = 10"},{name:"guidance_scale",val:": float = 0.0"},{name:"negative_prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"prompt_embeds_pooled",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds_pooled",val:": typing.Optional[torch.Tensor] = None"},{name:"num_images_per_prompt",val:": int = 1"},{name:"generator",val:": typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None"},{name:"latents",val:": typing.Optional[torch.Tensor] = None"},{name:"output_type",val:": typing.Optional[str] = 'pil'"},{name:"return_dict",val:": bool = True"},{name:"callback_on_step_end",val:": typing.Optional[typing.Callable[[int, int, typing.Dict], NoneType]] = None"},{name:"callback_on_step_end_tensor_inputs",val:": typing.List[str] = ['latents']"}],parametersDescription:[{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.image_embedding",description:`<strong>image_embedding</strong> (<code>torch.Tensor</code> or <code>List[torch.Tensor]</code>) &#x2014;
Image Embeddings either extracted from an image or generated by a Prior Model.`,name:"image_embedding"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>) &#x2014;
The prompt or prompts to guide the image generation.`,name:"prompt"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.num_inference_steps",description:`<strong>num_inference_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 12) &#x2014;
The number of denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference.`,name:"num_inference_steps"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.guidance_scale",description:`<strong>guidance_scale</strong> (<code>float</code>, <em>optional</em>, defaults to 0.0) &#x2014;
Guidance scale as defined in <a href="https://huggingface.co/papers/2207.12598" rel="nofollow">Classifier-Free Diffusion
Guidance</a>. <code>decoder_guidance_scale</code> is defined as <code>w</code> of
equation 2. of <a href="https://huggingface.co/papers/2205.11487" rel="nofollow">Imagen Paper</a>. Guidance scale is enabled by
setting <code>decoder_guidance_scale &gt; 1</code>. Higher guidance scale encourages to generate images that are
closely linked to the text <code>prompt</code>, usually at the expense of lower image quality.`,name:"guidance_scale"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.negative_prompt",description:`<strong>negative_prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
if <code>decoder_guidance_scale</code> is less than <code>1</code>).`,name:"negative_prompt"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.prompt_embeds",description:`<strong>prompt_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt weighting. If not
provided, text embeddings will be generated from <code>prompt</code> input argument.`,name:"prompt_embeds"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.prompt_embeds_pooled",description:`<strong>prompt_embeds_pooled</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt weighting.
If not provided, pooled text embeddings will be generated from <code>prompt</code> input argument.`,name:"prompt_embeds_pooled"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.negative_prompt_embeds",description:`<strong>negative_prompt_embeds</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt
weighting. If not provided, negative_prompt_embeds will be generated from <code>negative_prompt</code> input
argument.`,name:"negative_prompt_embeds"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.negative_prompt_embeds_pooled",description:`<strong>negative_prompt_embeds_pooled</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, <em>e.g.</em> prompt
weighting. If not provided, negative_prompt_embeds_pooled will be generated from <code>negative_prompt</code>
input argument.`,name:"negative_prompt_embeds_pooled"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.num_images_per_prompt",description:`<strong>num_images_per_prompt</strong> (<code>int</code>, <em>optional</em>, defaults to 1) &#x2014;
The number of images to generate per prompt.`,name:"num_images_per_prompt"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.generator",description:`<strong>generator</strong> (<code>torch.Generator</code> or <code>List[torch.Generator]</code>, <em>optional</em>) &#x2014;
One or a list of <a href="https://pytorch.org/docs/stable/generated/torch.Generator.html" rel="nofollow">torch generator(s)</a>
to make generation deterministic.`,name:"generator"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.latents",description:`<strong>latents</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will be generated by sampling using the supplied random <code>generator</code>.`,name:"latents"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.output_type",description:`<strong>output_type</strong> (<code>str</code>, <em>optional</em>, defaults to <code>&quot;pil&quot;</code>) &#x2014;
The output format of the generate image. Choose between: <code>&quot;pil&quot;</code> (<code>PIL.Image.Image</code>), <code>&quot;np&quot;</code>
(<code>np.array</code>) or <code>&quot;pt&quot;</code> (<code>torch.Tensor</code>).`,name:"output_type"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether or not to return a <a href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput">ImagePipelineOutput</a> instead of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.callback_on_step_end",description:`<strong>callback_on_step_end</strong> (<code>Callable</code>, <em>optional</em>) &#x2014;
A function that calls at the end of each denoising steps during the inference. The function is called
with the following arguments: <code>callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)</code>. <code>callback_kwargs</code> will include a list of all tensors as specified by
<code>callback_on_step_end_tensor_inputs</code>.`,name:"callback_on_step_end"},{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.callback_on_step_end_tensor_inputs",description:`<strong>callback_on_step_end_tensor_inputs</strong> (<code>List</code>, <em>optional</em>) &#x2014;
The list of tensor inputs for the <code>callback_on_step_end</code> function. The tensors specified in the list
will be passed as <code>callback_kwargs</code> argument. You will only be able to include variables listed in the
<code>._callback_tensor_inputs</code> attribute of your pipeline class.`,name:"callback_on_step_end_tensor_inputs"}],source:"https://github.com/huggingface/diffusers/blob/vr_12595/src/diffusers/pipelines/stable_cascade/pipeline_stable_cascade.py#L304",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


<p><a
  href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput"
>ImagePipelineOutput</a> or <code>tuple</code> <a
  href="/docs/diffusers/pr_12595/en/api/pipelines/ddim#diffusers.ImagePipelineOutput"
>ImagePipelineOutput</a> if <code>return_dict</code> is True,
otherwise a <code>tuple</code>. When returning a tuple, the first element is a list with the generated image
embeddings.</p>
`}}),L=new Ot({props:{anchor:"diffusers.StableCascadeDecoderPipeline.__call__.example",$$slots:{default:[Gn]},$$scope:{ctx:G}}}),Ie=new $n({props:{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/pipelines/stable_cascade.md"}}),{c(){_=r("meta"),J=o(),w=r("p"),b=o(),c(y.$$.fragment),s=o(),c(C.$$.fragment),Xe=o(),H=r("p"),H.innerHTML=en,ze=o(),Q=r("p"),Q.textContent=tn,Fe=o(),X=r("p"),X.innerHTML=nn,qe=o(),c(z.$$.fragment),Ae=o(),F=r("p"),F.textContent=an,Oe=o(),q=r("p"),q.textContent=on,Ke=o(),A=r("p"),A.innerHTML=sn,et=o(),O=r("p"),O.innerHTML=rn,tt=o(),B=r("blockquote"),B.innerHTML=ln,nt=o(),c(K.$$.fragment),at=o(),c(ee.$$.fragment),ot=o(),c(te.$$.fragment),it=o(),c(ne.$$.fragment),st=o(),c(ae.$$.fragment),rt=o(),oe=r("p"),oe.innerHTML=dn,lt=o(),c(ie.$$.fragment),dt=o(),c(se.$$.fragment),ct=o(),c(re.$$.fragment),pt=o(),le=r("p"),le.textContent=cn,mt=o(),de=r("ul"),de.innerHTML=pn,gt=o(),ce=r("p"),ce.textContent=mn,ut=o(),c(pe.$$.fragment),ft=o(),me=r("p"),me.innerHTML=gn,_t=o(),c(ge.$$.fragment),ht=o(),c(ue.$$.fragment),bt=o(),fe=r("ul"),fe.innerHTML=un,yt=o(),c(_e.$$.fragment),wt=o(),T=r("div"),c(he.$$.fragment),Zt=o(),Se=r("p"),Se.textContent=fn,$t=o(),Ze=r("p"),Ze.innerHTML=_n,kt=o(),j=r("div"),c(be.$$.fragment),Wt=o(),$e=r("p"),$e.textContent=hn,Gt=o(),c(N.$$.fragment),Bt=o(),V=r("div"),c(ye.$$.fragment),Nt=o(),ke=r("p"),ke.innerHTML=bn,Vt=o(),Y=r("div"),c(we.$$.fragment),Yt=o(),We=r("p"),We.innerHTML=yn,Ct=o(),c(Ce.$$.fragment),Tt=o(),U=r("div"),c(Te.$$.fragment),Rt=o(),Ge=r("p"),Ge.textContent=wn,Lt=o(),Be=r("p"),Be.innerHTML=Cn,Dt=o(),S=r("div"),c(Me.$$.fragment),Et=o(),Ne=r("p"),Ne.textContent=Tn,Ht=o(),c(R.$$.fragment),Mt=o(),c(ve.$$.fragment),vt=o(),$=r("div"),c(Je.$$.fragment),Qt=o(),Ve=r("p"),Ve.textContent=Mn,Jt=o(),c(Ue.$$.fragment),Ut=o(),P=r("div"),c(Pe.$$.fragment),Xt=o(),Ye=r("p"),Ye.textContent=vn,zt=o(),Re=r("p"),Re.innerHTML=Jn,Ft=o(),Z=r("div"),c(xe.$$.fragment),qt=o(),Le=r("p"),Le.textContent=Un,At=o(),c(L.$$.fragment),Pt=o(),c(Ie.$$.fragment),xt=o(),Qe=r("p"),this.h()},l(e){const t=Sn("svelte-u9bgzb",document.head);_=l(t,"META",{name:!0,content:!0}),t.forEach(n),J=i(e),w=l(e,"P",{}),I(w).forEach(n),b=i(e),p(y.$$.fragment,e),s=i(e),p(C.$$.fragment,e),Xe=i(e),H=l(e,"P",{"data-svelte-h":!0}),h(H)!=="svelte-58yed7"&&(H.innerHTML=en),ze=i(e),Q=l(e,"P",{"data-svelte-h":!0}),h(Q)!=="svelte-pdgmnb"&&(Q.textContent=tn),Fe=i(e),X=l(e,"P",{"data-svelte-h":!0}),h(X)!=="svelte-1rol617"&&(X.innerHTML=nn),qe=i(e),p(z.$$.fragment,e),Ae=i(e),F=l(e,"P",{"data-svelte-h":!0}),h(F)!=="svelte-13mtumf"&&(F.textContent=an),Oe=i(e),q=l(e,"P",{"data-svelte-h":!0}),h(q)!=="svelte-1lpwfdo"&&(q.textContent=on),Ke=i(e),A=l(e,"P",{"data-svelte-h":!0}),h(A)!=="svelte-nx9r15"&&(A.innerHTML=sn),et=i(e),O=l(e,"P",{"data-svelte-h":!0}),h(O)!=="svelte-1r2fmf1"&&(O.innerHTML=rn),tt=i(e),B=l(e,"BLOCKQUOTE",{class:!0,"data-svelte-h":!0}),h(B)!=="svelte-12eusja"&&(B.innerHTML=ln),nt=i(e),p(K.$$.fragment,e),at=i(e),p(ee.$$.fragment,e),ot=i(e),p(te.$$.fragment,e),it=i(e),p(ne.$$.fragment,e),st=i(e),p(ae.$$.fragment,e),rt=i(e),oe=l(e,"P",{"data-svelte-h":!0}),h(oe)!=="svelte-vq9k87"&&(oe.innerHTML=dn),lt=i(e),p(ie.$$.fragment,e),dt=i(e),p(se.$$.fragment,e),ct=i(e),p(re.$$.fragment,e),pt=i(e),le=l(e,"P",{"data-svelte-h":!0}),h(le)!=="svelte-marzm0"&&(le.textContent=cn),mt=i(e),de=l(e,"UL",{"data-svelte-h":!0}),h(de)!=="svelte-1io3utn"&&(de.innerHTML=pn),gt=i(e),ce=l(e,"P",{"data-svelte-h":!0}),h(ce)!=="svelte-usmk6k"&&(ce.textContent=mn),ut=i(e),p(pe.$$.fragment,e),ft=i(e),me=l(e,"P",{"data-svelte-h":!0}),h(me)!=="svelte-md3bci"&&(me.innerHTML=gn),_t=i(e),p(ge.$$.fragment,e),ht=i(e),p(ue.$$.fragment,e),bt=i(e),fe=l(e,"UL",{"data-svelte-h":!0}),h(fe)!=="svelte-1lgsa2f"&&(fe.innerHTML=un),yt=i(e),p(_e.$$.fragment,e),wt=i(e),T=l(e,"DIV",{class:!0});var M=I(T);p(he.$$.fragment,M),Zt=i(M),Se=l(M,"P",{"data-svelte-h":!0}),h(Se)!=="svelte-zuq8la"&&(Se.textContent=fn),$t=i(M),Ze=l(M,"P",{"data-svelte-h":!0}),h(Ze)!=="svelte-1mudibp"&&(Ze.innerHTML=_n),kt=i(M),j=l(M,"DIV",{class:!0});var k=I(j);p(be.$$.fragment,k),Wt=i(k),$e=l(k,"P",{"data-svelte-h":!0}),h($e)!=="svelte-v78lg8"&&($e.textContent=hn),Gt=i(k),p(N.$$.fragment,k),k.forEach(n),Bt=i(M),V=l(M,"DIV",{class:!0});var je=I(V);p(ye.$$.fragment,je),Nt=i(je),ke=l(je,"P",{"data-svelte-h":!0}),h(ke)!=="svelte-d7vywi"&&(ke.innerHTML=bn),je.forEach(n),Vt=i(M),Y=l(M,"DIV",{class:!0});var jt=I(Y);p(we.$$.fragment,jt),Yt=i(jt),We=l(jt,"P",{"data-svelte-h":!0}),h(We)!=="svelte-a2a0wt"&&(We.innerHTML=yn),jt.forEach(n),M.forEach(n),Ct=i(e),p(Ce.$$.fragment,e),Tt=i(e),U=l(e,"DIV",{class:!0});var D=I(U);p(Te.$$.fragment,D),Rt=i(D),Ge=l(D,"P",{"data-svelte-h":!0}),h(Ge)!=="svelte-1k7c4ws"&&(Ge.textContent=wn),Lt=i(D),Be=l(D,"P",{"data-svelte-h":!0}),h(Be)!=="svelte-1mudibp"&&(Be.innerHTML=Cn),Dt=i(D),S=l(D,"DIV",{class:!0});var De=I(S);p(Me.$$.fragment,De),Et=i(De),Ne=l(De,"P",{"data-svelte-h":!0}),h(Ne)!=="svelte-v78lg8"&&(Ne.textContent=Tn),Ht=i(De),p(R.$$.fragment,De),De.forEach(n),D.forEach(n),Mt=i(e),p(ve.$$.fragment,e),vt=i(e),$=l(e,"DIV",{class:!0});var St=I($);p(Je.$$.fragment,St),Qt=i(St),Ve=l(St,"P",{"data-svelte-h":!0}),h(Ve)!=="svelte-cw9h8a"&&(Ve.textContent=Mn),St.forEach(n),Jt=i(e),p(Ue.$$.fragment,e),Ut=i(e),P=l(e,"DIV",{class:!0});var E=I(P);p(Pe.$$.fragment,E),Xt=i(E),Ye=l(E,"P",{"data-svelte-h":!0}),h(Ye)!=="svelte-1bqijj8"&&(Ye.textContent=vn),zt=i(E),Re=l(E,"P",{"data-svelte-h":!0}),h(Re)!=="svelte-1mudibp"&&(Re.innerHTML=Jn),Ft=i(E),Z=l(E,"DIV",{class:!0});var Ee=I(Z);p(xe.$$.fragment,Ee),qt=i(Ee),Le=l(Ee,"P",{"data-svelte-h":!0}),h(Le)!=="svelte-v78lg8"&&(Le.textContent=Un),At=i(Ee),p(L.$$.fragment,Ee),Ee.forEach(n),E.forEach(n),Pt=i(e),p(Ie.$$.fragment,e),xt=i(e),Qe=l(e,"P",{}),I(Qe).forEach(n),this.h()},h(){x(_,"name","hf:doc:metadata"),x(_,"content",Nn),x(B,"class","warning"),x(j,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(V,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(Y,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(T,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(S,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(U,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x($,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(Z,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),x(P,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8")},m(e,t){d(document.head,_),a(e,J,t),a(e,w,t),a(e,b,t),m(y,e,t),a(e,s,t),m(C,e,t),a(e,Xe,t),a(e,H,t),a(e,ze,t),a(e,Q,t),a(e,Fe,t),a(e,X,t),a(e,qe,t),m(z,e,t),a(e,Ae,t),a(e,F,t),a(e,Oe,t),a(e,q,t),a(e,Ke,t),a(e,A,t),a(e,et,t),a(e,O,t),a(e,tt,t),a(e,B,t),a(e,nt,t),m(K,e,t),a(e,at,t),m(ee,e,t),a(e,ot,t),m(te,e,t),a(e,it,t),m(ne,e,t),a(e,st,t),m(ae,e,t),a(e,rt,t),a(e,oe,t),a(e,lt,t),m(ie,e,t),a(e,dt,t),m(se,e,t),a(e,ct,t),m(re,e,t),a(e,pt,t),a(e,le,t),a(e,mt,t),a(e,de,t),a(e,gt,t),a(e,ce,t),a(e,ut,t),m(pe,e,t),a(e,ft,t),a(e,me,t),a(e,_t,t),m(ge,e,t),a(e,ht,t),m(ue,e,t),a(e,bt,t),a(e,fe,t),a(e,yt,t),m(_e,e,t),a(e,wt,t),a(e,T,t),m(he,T,null),d(T,Zt),d(T,Se),d(T,$t),d(T,Ze),d(T,kt),d(T,j),m(be,j,null),d(j,Wt),d(j,$e),d(j,Gt),m(N,j,null),d(T,Bt),d(T,V),m(ye,V,null),d(V,Nt),d(V,ke),d(T,Vt),d(T,Y),m(we,Y,null),d(Y,Yt),d(Y,We),a(e,Ct,t),m(Ce,e,t),a(e,Tt,t),a(e,U,t),m(Te,U,null),d(U,Rt),d(U,Ge),d(U,Lt),d(U,Be),d(U,Dt),d(U,S),m(Me,S,null),d(S,Et),d(S,Ne),d(S,Ht),m(R,S,null),a(e,Mt,t),m(ve,e,t),a(e,vt,t),a(e,$,t),m(Je,$,null),d($,Qt),d($,Ve),a(e,Jt,t),m(Ue,e,t),a(e,Ut,t),a(e,P,t),m(Pe,P,null),d(P,Xt),d(P,Ye),d(P,zt),d(P,Re),d(P,Ft),d(P,Z),m(xe,Z,null),d(Z,qt),d(Z,Le),d(Z,At),m(L,Z,null),a(e,Pt,t),m(Ie,e,t),a(e,xt,t),a(e,Qe,t),It=!0},p(e,[t]){const M={};t&2&&(M.$$scope={dirty:t,ctx:e}),N.$set(M);const k={};t&2&&(k.$$scope={dirty:t,ctx:e}),R.$set(k);const je={};t&2&&(je.$$scope={dirty:t,ctx:e}),L.$set(je)},i(e){It||(g(y.$$.fragment,e),g(C.$$.fragment,e),g(z.$$.fragment,e),g(K.$$.fragment,e),g(ee.$$.fragment,e),g(te.$$.fragment,e),g(ne.$$.fragment,e),g(ae.$$.fragment,e),g(ie.$$.fragment,e),g(se.$$.fragment,e),g(re.$$.fragment,e),g(pe.$$.fragment,e),g(ge.$$.fragment,e),g(ue.$$.fragment,e),g(_e.$$.fragment,e),g(he.$$.fragment,e),g(be.$$.fragment,e),g(N.$$.fragment,e),g(ye.$$.fragment,e),g(we.$$.fragment,e),g(Ce.$$.fragment,e),g(Te.$$.fragment,e),g(Me.$$.fragment,e),g(R.$$.fragment,e),g(ve.$$.fragment,e),g(Je.$$.fragment,e),g(Ue.$$.fragment,e),g(Pe.$$.fragment,e),g(xe.$$.fragment,e),g(L.$$.fragment,e),g(Ie.$$.fragment,e),It=!0)},o(e){u(y.$$.fragment,e),u(C.$$.fragment,e),u(z.$$.fragment,e),u(K.$$.fragment,e),u(ee.$$.fragment,e),u(te.$$.fragment,e),u(ne.$$.fragment,e),u(ae.$$.fragment,e),u(ie.$$.fragment,e),u(se.$$.fragment,e),u(re.$$.fragment,e),u(pe.$$.fragment,e),u(ge.$$.fragment,e),u(ue.$$.fragment,e),u(_e.$$.fragment,e),u(he.$$.fragment,e),u(be.$$.fragment,e),u(N.$$.fragment,e),u(ye.$$.fragment,e),u(we.$$.fragment,e),u(Ce.$$.fragment,e),u(Te.$$.fragment,e),u(Me.$$.fragment,e),u(R.$$.fragment,e),u(ve.$$.fragment,e),u(Je.$$.fragment,e),u(Ue.$$.fragment,e),u(Pe.$$.fragment,e),u(xe.$$.fragment,e),u(L.$$.fragment,e),u(Ie.$$.fragment,e),It=!1},d(e){e&&(n(J),n(w),n(b),n(s),n(Xe),n(H),n(ze),n(Q),n(Fe),n(X),n(qe),n(Ae),n(F),n(Oe),n(q),n(Ke),n(A),n(et),n(O),n(tt),n(B),n(nt),n(at),n(ot),n(it),n(st),n(rt),n(oe),n(lt),n(dt),n(ct),n(pt),n(le),n(mt),n(de),n(gt),n(ce),n(ut),n(ft),n(me),n(_t),n(ht),n(bt),n(fe),n(yt),n(wt),n(T),n(Ct),n(Tt),n(U),n(Mt),n(vt),n($),n(Jt),n(Ut),n(P),n(Pt),n(xt),n(Qe)),n(_),f(y,e),f(C,e),f(z,e),f(K,e),f(ee,e),f(te,e),f(ne,e),f(ae,e),f(ie,e),f(se,e),f(re,e),f(pe,e),f(ge,e),f(ue,e),f(_e,e),f(he),f(be),f(N),f(ye),f(we),f(Ce,e),f(Te),f(Me),f(R),f(ve,e),f(Je),f(Ue,e),f(Pe),f(xe),f(L),f(Ie,e)}}}const Nn='{"title":"Stable Cascade","local":"stable-cascade","sections":[{"title":"Model Overview","local":"model-overview","sections":[],"depth":2},{"title":"Usage example","local":"usage-example","sections":[],"depth":2},{"title":"Using the Lite Versions of the Stage B and Stage C models","local":"using-the-lite-versions-of-the-stage-b-and-stage-c-models","sections":[],"depth":2},{"title":"Loading original checkpoints with from_single_file","local":"loading-original-checkpoints-with-fromsinglefile","sections":[],"depth":2},{"title":"Uses","local":"uses","sections":[{"title":"Direct Use","local":"direct-use","sections":[],"depth":3},{"title":"Out-of-Scope Use","local":"out-of-scope-use","sections":[],"depth":3}],"depth":2},{"title":"Limitations and Bias","local":"limitations-and-bias","sections":[{"title":"Limitations","local":"limitations","sections":[],"depth":3}],"depth":2},{"title":"StableCascadeCombinedPipeline","local":"diffusers.StableCascadeCombinedPipeline","sections":[],"depth":2},{"title":"StableCascadePriorPipeline","local":"diffusers.StableCascadePriorPipeline","sections":[],"depth":2},{"title":"StableCascadePriorPipelineOutput","local":"diffusers.pipelines.stable_cascade.pipeline_stable_cascade_prior.StableCascadePriorPipelineOutput","sections":[],"depth":2},{"title":"StableCascadeDecoderPipeline","local":"diffusers.StableCascadeDecoderPipeline","sections":[],"depth":2}],"depth":1}';function Vn(G){return xn(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class Xn extends In{constructor(_){super(),jn(this,_,Vn,Bn,Pn,{})}}export{Xn as component};