diffusers/pr_11783/en/_app/immutable/nodes/166.f2d516a5.js

import{s as io,o as ao,n as Tt}from"../chunks/scheduler.8c3d61f6.js";import{S as ro,i as lo,g as r,s,r as m,A as po,h as l,f as o,c as i,j as W,u as f,x as d,k as $,y as b,a as n,v as u,d as g,t as h,w as _}from"../chunks/index.da70eac4.js";import{T as no}from"../chunks/Tip.1d9b8c37.js";import{D as Ze}from"../chunks/Docstring.dcbc19b1.js";import{C as Je}from"../chunks/CodeBlock.a9c4becf.js";import{E as so}from"../chunks/ExampleCodeBlock.da4c0768.js";import{H as P,E as co}from"../chunks/getInferenceSnippets.86beaec3.js";function mo(J){let a,y="🧪 This pipeline is for research purposes only.";return{c(){a=r("p"),a.textContent=y},l(c){a=l(c,"P",{"data-svelte-h":!0}),d(a)!=="svelte-1oxhjjd"&&(a.textContent=y)},m(c,v){n(c,a,v)},p:Tt,d(c){c&&o(a)}}}function fo(J){let a,y='Make sure to check out the Schedulers <a href="../../using-diffusers/schedulers">guide</a> to learn how to explore the tradeoff between scheduler speed and quality, and see the <a href="../../using-diffusers/loading#reuse-a-pipeline">reuse components across pipelines</a> section to learn how to efficiently load the same components into multiple pipelines.';return{c(){a=r("p"),a.innerHTML=y},l(c){a=l(c,"P",{"data-svelte-h":!0}),d(a)!=="svelte-1qn15hi"&&(a.innerHTML=y)},m(c,v){n(c,a,v)},p:Tt,d(c){c&&o(a)}}}function uo(J){let a,y="Examples:",c,v,T;return v=new Je({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> TextToVideoSDPipeline
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_video

<span class="hljs-meta">&gt;&gt;&gt; </span>pipe = TextToVideoSDPipeline.from_pretrained(
<span class="hljs-meta">... </span>    <span class="hljs-string">&quot;damo-vilab/text-to-video-ms-1.7b&quot;</span>, torch_dtype=torch.float16, variant=<span class="hljs-string">&quot;fp16&quot;</span>
<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;Spiderman is surfing&quot;</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>video_frames = pipe(prompt).frames[<span class="hljs-number">0</span>]
<span class="hljs-meta">&gt;&gt;&gt; </span>video_path = export_to_video(video_frames)
<span class="hljs-meta">&gt;&gt;&gt; </span>video_path`,wrap:!1}}),{c(){a=r("p"),a.textContent=y,c=s(),m(v.$$.fragment)},l(p){a=l(p,"P",{"data-svelte-h":!0}),d(a)!=="svelte-kvfsh7"&&(a.textContent=y),c=i(p),f(v.$$.fragment,p)},m(p,w){n(p,a,w),n(p,c,w),u(v,p,w),T=!0},p:Tt,i(p){T||(g(v.$$.fragment,p),T=!0)},o(p){h(v.$$.fragment,p),T=!1},d(p){p&&(o(a),o(c)),_(v,p)}}}function go(J){let a,y="Examples:",c,v,T;return v=new Je({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> DiffusionPipeline, DPMSolverMultistepScheduler
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_video

<span class="hljs-meta">&gt;&gt;&gt; </span>pipe = DiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;cerspense/zeroscope_v2_576w&quot;</span>, torch_dtype=torch.float16)
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)

<span class="hljs-meta">&gt;&gt;&gt; </span>prompt = <span class="hljs-string">&quot;spiderman running in the desert&quot;</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>video_frames = pipe(prompt, num_inference_steps=<span class="hljs-number">40</span>, height=<span class="hljs-number">320</span>, width=<span class="hljs-number">576</span>, num_frames=<span class="hljs-number">24</span>).frames[<span class="hljs-number">0</span>]
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># safe low-res video</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>video_path = export_to_video(video_frames, output_video_path=<span class="hljs-string">&quot;./video_576_spiderman.mp4&quot;</span>)

<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># let&#x27;s offload the text-to-image model</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.to(<span class="hljs-string">&quot;cpu&quot;</span>)

<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># and load the image-to-image model</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe = DiffusionPipeline.from_pretrained(
<span class="hljs-meta">... </span>    <span class="hljs-string">&quot;cerspense/zeroscope_v2_XL&quot;</span>, torch_dtype=torch.float16, revision=<span class="hljs-string">&quot;refs/pr/15&quot;</span>
<span class="hljs-meta">... </span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.enable_model_cpu_offload()

<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># The VAE consumes A LOT of memory, let&#x27;s make sure we run it in sliced mode</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>pipe.vae.enable_slicing()

<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># now let&#x27;s upscale it</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>video = [Image.fromarray(frame).resize((<span class="hljs-number">1024</span>, <span class="hljs-number">576</span>)) <span class="hljs-keyword">for</span> frame <span class="hljs-keyword">in</span> video_frames]

<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># and denoise it</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>video_frames = pipe(prompt, video=video, strength=<span class="hljs-number">0.6</span>).frames[<span class="hljs-number">0</span>]
<span class="hljs-meta">&gt;&gt;&gt; </span>video_path = export_to_video(video_frames, output_video_path=<span class="hljs-string">&quot;./video_1024_spiderman.mp4&quot;</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>video_path`,wrap:!1}}),{c(){a=r("p"),a.textContent=y,c=s(),m(v.$$.fragment)},l(p){a=l(p,"P",{"data-svelte-h":!0}),d(a)!=="svelte-kvfsh7"&&(a.textContent=y),c=i(p),f(v.$$.fragment,p)},m(p,w){n(p,a,w),n(p,c,w),u(v,p,w),T=!0},p:Tt,i(p){T||(g(v.$$.fragment,p),T=!0)},o(p){h(v.$$.fragment,p),T=!1},d(p){p&&(o(a),o(c)),_(v,p)}}}function ho(J){let a,y,c,v,T,p,w,Ce,I,Wt='<img alt="LoRA" src="https://img.shields.io/badge/LoRA-d8b4fe?style=flat"/>',De,R,It='<a href="https://huggingface.co/papers/2308.06571" rel="nofollow">ModelScope Text-to-Video Technical Report</a> is by Jiuniu Wang, Hangjie Yuan, Dayou Chen, Yingya Zhang, Xiang Wang, Shiwei Zhang.',Be,L,St="The abstract from the paper is:",Xe,z,Ct='<em>This paper introduces ModelScopeT2V, a text-to-video synthesis model that evolves from a text-to-image synthesis model (i.e., Stable Diffusion). ModelScopeT2V incorporates spatio-temporal blocks to ensure consistent frame generation and smooth movement transitions. The model could adapt to varying frame numbers during training and inference, rendering it suitable for both image-text and video-text datasets. ModelScopeT2V brings together three components (i.e., VQGAN, a text encoder, and a denoising UNet), totally comprising 1.7 billion parameters, in which 0.5 billion parameters are dedicated to temporal capabilities. The model demonstrates superior performance over state-of-the-art methods across three evaluation metrics. The code and an online demo are available at <a href="https://modelscope.cn/models/damo/text-to-video-synthesis/summary" rel="nofollow">https://modelscope.cn/models/damo/text-to-video-synthesis/summary</a>.</em>',Pe,N,Dt='You can find additional information about Text-to-Video on the <a href="https://modelscope.cn/models/damo/text-to-video-synthesis/summary" rel="nofollow">project page</a>, <a href="https://github.com/modelscope/modelscope/" rel="nofollow">original codebase</a>, and try it out in a <a href="https://huggingface.co/spaces/damo-vilab/modelscope-text-to-video-synthesis" rel="nofollow">demo</a>. Official checkpoints can be found at <a href="https://huggingface.co/damo-vilab" rel="nofollow">damo-vilab</a> and <a href="https://huggingface.co/cerspense" rel="nofollow">cerspense</a>.',Re,E,Le,Y,ze,H,Bt="Let’s start by generating a short video with the default length of 16 frames (2s at 8 fps):",Ne,F,Ee,Q,Xt=`Diffusers supports different optimization techniques to improve the latency
and memory footprint of a pipeline. Since videos are often more memory-heavy than images,
we can enable CPU offloading and VAE slicing to keep the memory footprint at bay.`,Ye,q,Pt="Let’s generate a video of 8 seconds (64 frames) on the same GPU using CPU offloading and VAE slicing:",He,A,Fe,O,Rt="It just takes <strong>7 GBs of GPU memory</strong> to generate the 64 video frames using PyTorch 2.0, “fp16” precision and the techniques mentioned above.",Qe,K,Lt="We can also use a different scheduler easily, using the same method we’d use for Stable Diffusion:",qe,ee,Ae,te,zt="Here are some sample outputs:",Oe,oe,Nt=`<tbody><tr><td><center>An astronaut riding a horse.
        <br/> <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astr.gif" alt="An astronaut riding a horse." style="width: 300px;"/></center></td> <td><center>Darth vader surfing in waves.
        <br/> <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/vader.gif" alt="Darth vader surfing in waves." style="width: 300px;"/></center></td></tr></tbody>`,Ke,ne,et,se,Et=`Zeroscope are watermark-free model and have been trained on specific sizes such as <code>576x320</code> and <code>1024x576</code>.
One should first generate a video using the lower resolution checkpoint <a href="https://huggingface.co/cerspense/zeroscope_v2_576w" rel="nofollow"><code>cerspense/zeroscope_v2_576w</code></a> with <a href="/docs/diffusers/pr_11783/en/api/pipelines/text_to_video#diffusers.TextToVideoSDPipeline">TextToVideoSDPipeline</a>,
which can then be upscaled using <a href="/docs/diffusers/pr_11783/en/api/pipelines/text_to_video#diffusers.VideoToVideoSDPipeline">VideoToVideoSDPipeline</a> and <a href="https://huggingface.co/cerspense/zeroscope_v2_XL" rel="nofollow"><code>cerspense/zeroscope_v2_XL</code></a>.`,tt,ie,ot,ae,Yt="Now the video can be upscaled:",nt,re,st,le,Ht="Here are some sample outputs:",it,pe,Ft=`<tbody><tr><td><center>Darth vader surfing in waves.
        <br/> <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/darthvader_cerpense.gif" alt="Darth vader surfing in waves." style="width: 576px;"/></center></td></tr></tbody>`,at,de,rt,ce,Qt="Video generation is memory-intensive and one way to reduce your memory usage is to set <code>enable_forward_chunking</code> on the pipeline’s UNet so you don’t run the entire feedforward layer at once. Breaking it up into chunks in a loop is more efficient.",lt,me,qt='Check out the <a href="text-img2vid">Text or image-to-video</a> guide for more details about how certain parameters can affect video generation and how to optimize inference by reducing memory usage.',pt,S,dt,fe,ct,x,ue,yt,V,ge,wt,Ve,At="The call function to the pipeline for generation.",xt,C,Mt,D,he,Zt,je,Ot="Encodes the prompt into text encoder hidden states.",mt,_e,ft,M,ve,Jt,j,be,Vt,$e,Kt="The call function to the pipeline for generation.",jt,B,$t,X,Te,kt,ke,eo="Encodes the prompt into text encoder hidden states.",ut,ye,gt,Z,we,Gt,Ge,to="Output class for text-to-video pipelines.",Ut,Ue,oo=`PIL image sequences of length <code>num_frames.</code> It can also be a NumPy array or Torch tensor of shape
<code>(batch_size, num_frames, channels, height, width)</code>`,ht,xe,_t,Se,vt;return T=new no({props:{warning:!0,$$slots:{default:[mo]},$$scope:{ctx:J}}}),w=new P({props:{title:"Text-to-video",local:"text-to-video",headingTag:"h1"}}),E=new P({props:{title:"Usage example",local:"usage-example",headingTag:"h2"}}),Y=new P({props:{title:"text-to-video-ms-1.7b",local:"text-to-video-ms-17b",headingTag:"h3"}}),F=new Je({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> DiffusionPipeline
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_video

pipe = DiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;damo-vilab/text-to-video-ms-1.7b&quot;</span>, torch_dtype=torch.float16, variant=<span class="hljs-string">&quot;fp16&quot;</span>)
pipe = pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)

prompt = <span class="hljs-string">&quot;Spiderman is surfing&quot;</span>
video_frames = pipe(prompt).frames[<span class="hljs-number">0</span>]
video_path = export_to_video(video_frames)
video_path`,wrap:!1}}),A=new Je({props:{code:"aW1wb3J0JTIwdG9yY2glMEFmcm9tJTIwZGlmZnVzZXJzJTIwaW1wb3J0JTIwRGlmZnVzaW9uUGlwZWxpbmUlMEFmcm9tJTIwZGlmZnVzZXJzLnV0aWxzJTIwaW1wb3J0JTIwZXhwb3J0X3RvX3ZpZGVvJTBBJTBBcGlwZSUyMCUzRCUyMERpZmZ1c2lvblBpcGVsaW5lLmZyb21fcHJldHJhaW5lZCglMjJkYW1vLXZpbGFiJTJGdGV4dC10by12aWRlby1tcy0xLjdiJTIyJTJDJTIwdG9yY2hfZHR5cGUlM0R0b3JjaC5mbG9hdDE2JTJDJTIwdmFyaWFudCUzRCUyMmZwMTYlMjIpJTBBcGlwZS5lbmFibGVfbW9kZWxfY3B1X29mZmxvYWQoKSUwQSUwQSUyMyUyMG1lbW9yeSUyMG9wdGltaXphdGlvbiUwQXBpcGUuZW5hYmxlX3ZhZV9zbGljaW5nKCklMEElMEFwcm9tcHQlMjAlM0QlMjAlMjJEYXJ0aCUyMFZhZGVyJTIwc3VyZmluZyUyMGElMjB3YXZlJTIyJTBBdmlkZW9fZnJhbWVzJTIwJTNEJTIwcGlwZShwcm9tcHQlMkMlMjBudW1fZnJhbWVzJTNENjQpLmZyYW1lcyU1QjAlNUQlMEF2aWRlb19wYXRoJTIwJTNEJTIwZXhwb3J0X3RvX3ZpZGVvKHZpZGVvX2ZyYW1lcyklMEF2aWRlb19wYXRo",highlighted:`<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_video

pipe = DiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;damo-vilab/text-to-video-ms-1.7b&quot;</span>, torch_dtype=torch.float16, variant=<span class="hljs-string">&quot;fp16&quot;</span>)
pipe.enable_model_cpu_offload()

<span class="hljs-comment"># memory optimization</span>
pipe.enable_vae_slicing()

prompt = <span class="hljs-string">&quot;Darth Vader surfing a wave&quot;</span>
video_frames = pipe(prompt, num_frames=<span class="hljs-number">64</span>).frames[<span class="hljs-number">0</span>]
video_path = export_to_video(video_frames)
video_path`,wrap:!1}}),ee=new Je({props:{code:"aW1wb3J0JTIwdG9yY2glMEFmcm9tJTIwZGlmZnVzZXJzJTIwaW1wb3J0JTIwRGlmZnVzaW9uUGlwZWxpbmUlMkMlMjBEUE1Tb2x2ZXJNdWx0aXN0ZXBTY2hlZHVsZXIlMEFmcm9tJTIwZGlmZnVzZXJzLnV0aWxzJTIwaW1wb3J0JTIwZXhwb3J0X3RvX3ZpZGVvJTBBJTBBcGlwZSUyMCUzRCUyMERpZmZ1c2lvblBpcGVsaW5lLmZyb21fcHJldHJhaW5lZCglMjJkYW1vLXZpbGFiJTJGdGV4dC10by12aWRlby1tcy0xLjdiJTIyJTJDJTIwdG9yY2hfZHR5cGUlM0R0b3JjaC5mbG9hdDE2JTJDJTIwdmFyaWFudCUzRCUyMmZwMTYlMjIpJTBBcGlwZS5zY2hlZHVsZXIlMjAlM0QlMjBEUE1Tb2x2ZXJNdWx0aXN0ZXBTY2hlZHVsZXIuZnJvbV9jb25maWcocGlwZS5zY2hlZHVsZXIuY29uZmlnKSUwQXBpcGUuZW5hYmxlX21vZGVsX2NwdV9vZmZsb2FkKCklMEElMEFwcm9tcHQlMjAlM0QlMjAlMjJTcGlkZXJtYW4lMjBpcyUyMHN1cmZpbmclMjIlMEF2aWRlb19mcmFtZXMlMjAlM0QlMjBwaXBlKHByb21wdCUyQyUyMG51bV9pbmZlcmVuY2Vfc3RlcHMlM0QyNSkuZnJhbWVzJTVCMCU1RCUwQXZpZGVvX3BhdGglMjAlM0QlMjBleHBvcnRfdG9fdmlkZW8odmlkZW9fZnJhbWVzKSUwQXZpZGVvX3BhdGg=",highlighted:`<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, DPMSolverMultistepScheduler
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_video

pipe = DiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;damo-vilab/text-to-video-ms-1.7b&quot;</span>, torch_dtype=torch.float16, variant=<span class="hljs-string">&quot;fp16&quot;</span>)
pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
pipe.enable_model_cpu_offload()

prompt = <span class="hljs-string">&quot;Spiderman is surfing&quot;</span>
video_frames = pipe(prompt, num_inference_steps=<span class="hljs-number">25</span>).frames[<span class="hljs-number">0</span>]
video_path = export_to_video(video_frames)
video_path`,wrap:!1}}),ne=new P({props:{title:"cerspense/zeroscope_v2_576w & cerspense/zeroscope_v2_XL",local:"cerspensezeroscopev2576w--cerspensezeroscopev2xl",headingTag:"h3"}}),ie=new Je({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> DiffusionPipeline, DPMSolverMultistepScheduler
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_video
<span class="hljs-keyword">from</span> PIL <span class="hljs-keyword">import</span> Image

pipe = DiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;cerspense/zeroscope_v2_576w&quot;</span>, torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()

<span class="hljs-comment"># memory optimization</span>
pipe.unet.enable_forward_chunking(chunk_size=<span class="hljs-number">1</span>, dim=<span class="hljs-number">1</span>)
pipe.enable_vae_slicing()

prompt = <span class="hljs-string">&quot;Darth Vader surfing a wave&quot;</span>
video_frames = pipe(prompt, num_frames=<span class="hljs-number">24</span>).frames[<span class="hljs-number">0</span>]
video_path = export_to_video(video_frames)
video_path`,wrap:!1}}),re=new Je({props:{code:"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",highlighted:`pipe = DiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;cerspense/zeroscope_v2_XL&quot;</span>, torch_dtype=torch.float16)
pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
pipe.enable_model_cpu_offload()

<span class="hljs-comment"># memory optimization</span>
pipe.unet.enable_forward_chunking(chunk_size=<span class="hljs-number">1</span>, dim=<span class="hljs-number">1</span>)
pipe.enable_vae_slicing()

video = [Image.fromarray(frame).resize((<span class="hljs-number">1024</span>, <span class="hljs-number">576</span>)) <span class="hljs-keyword">for</span> frame <span class="hljs-keyword">in</span> video_frames]

video_frames = pipe(prompt, video=video, strength=<span class="hljs-number">0.6</span>).frames[<span class="hljs-number">0</span>]
video_path = export_to_video(video_frames)
video_path`,wrap:!1}}),de=new P({props:{title:"Tips",local:"tips",headingTag:"h2"}}),S=new no({props:{$$slots:{default:[fo]},$$scope:{ctx:J}}}),fe=new P({props:{title:"TextToVideoSDPipeline",local:"diffusers.TextToVideoSDPipeline",headingTag:"h2"}}),ue=new Ze({props:{name:"class diffusers.TextToVideoSDPipeline",anchor:"diffusers.TextToVideoSDPipeline",parameters:[{name:"vae",val:": AutoencoderKL"},{name:"text_encoder",val:": CLIPTextModel"},{name:"tokenizer",val:": CLIPTokenizer"},{name:"unet",val:": UNet3DConditionModel"},{name:"scheduler",val:": KarrasDiffusionSchedulers"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth.py#L70"}}),ge=new Ze({props:{name:"__call__",anchor:"diffusers.TextToVideoSDPipeline.__call__",parameters:[{name:"prompt",val:": typing.Union[str, typing.List[str]] = None"},{name:"height",val:": typing.Optional[int] = None"},{name:"width",val:": typing.Optional[int] = None"},{name:"num_frames",val:": int = 16"},{name:"num_inference_steps",val:": int = 50"},{name:"guidance_scale",val:": float = 9.0"},{name:"negative_prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"eta",val:": float = 0.0"},{name:"generator",val:": typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None"},{name:"latents",val:": typing.Optional[torch.Tensor] = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"output_type",val:": typing.Optional[str] = 'np'"},{name:"return_dict",val:": bool = True"},{name:"callback",val:": typing.Optional[typing.Callable[[int, int, torch.Tensor], NoneType]] = None"},{name:"callback_steps",val:": int = 1"},{name:"cross_attention_kwargs",val:": typing.Optional[typing.Dict[str, typing.Any]] = None"},{name:"clip_skip",val:": typing.Optional[int] = None"}],parametersDescription:[{anchor:"diffusers.TextToVideoSDPipeline.__call__.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts to guide image generation. If not defined, you need to pass <code>prompt_embeds</code>.`,name:"prompt"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.height",description:`<strong>height</strong> (<code>int</code>, <em>optional</em>, defaults to <code>self.unet.config.sample_size * self.vae_scale_factor</code>) &#x2014;
The height in pixels of the generated video.`,name:"height"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.width",description:`<strong>width</strong> (<code>int</code>, <em>optional</em>, defaults to <code>self.unet.config.sample_size * self.vae_scale_factor</code>) &#x2014;
The width in pixels of the generated video.`,name:"width"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.num_frames",description:`<strong>num_frames</strong> (<code>int</code>, <em>optional</em>, defaults to 16) &#x2014;
The number of video frames that are generated. Defaults to 16 frames which at 8 frames per seconds
amounts to 2 seconds of video.`,name:"num_frames"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.num_inference_steps",description:`<strong>num_inference_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 50) &#x2014;
The number of denoising steps. More denoising steps usually lead to a higher quality videos at the
expense of slower inference.`,name:"num_inference_steps"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.guidance_scale",description:`<strong>guidance_scale</strong> (<code>float</code>, <em>optional</em>, defaults to 7.5) &#x2014;
A higher guidance scale value encourages the model to generate images closely linked to the text
<code>prompt</code> at the expense of lower image quality. Guidance scale is enabled when <code>guidance_scale &gt; 1</code>.`,name:"guidance_scale"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.negative_prompt",description:`<strong>negative_prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts to guide what to not include in image generation. If not defined, you need to
pass <code>negative_prompt_embeds</code> instead. Ignored when not using guidance (<code>guidance_scale &lt; 1</code>).`,name:"negative_prompt"},{anchor:"diffusers.TextToVideoSDPipeline.__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.TextToVideoSDPipeline.__call__.eta",description:`<strong>eta</strong> (<code>float</code>, <em>optional</em>, defaults to 0.0) &#x2014;
Corresponds to parameter eta (&#x3B7;) from the <a href="https://huggingface.co/papers/2010.02502" rel="nofollow">DDIM</a> paper. Only
applies to the <a href="/docs/diffusers/pr_11783/en/api/schedulers/ddim#diffusers.DDIMScheduler">DDIMScheduler</a>, and is ignored in other schedulers.`,name:"eta"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.generator",description:`<strong>generator</strong> (<code>torch.Generator</code> or <code>List[torch.Generator]</code>, <em>optional</em>) &#x2014;
A <a href="https://pytorch.org/docs/stable/generated/torch.Generator.html" rel="nofollow"><code>torch.Generator</code></a> to make
generation deterministic.`,name:"generator"},{anchor:"diffusers.TextToVideoSDPipeline.__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 video
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random <code>generator</code>. Latents should be of shape
<code>(batch_size, num_channel, num_frames, height, width)</code>.`,name:"latents"},{anchor:"diffusers.TextToVideoSDPipeline.__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 (prompt weighting). If not
provided, text embeddings are generated from the <code>prompt</code> input argument.`,name:"prompt_embeds"},{anchor:"diffusers.TextToVideoSDPipeline.__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 (prompt weighting). If
not provided, <code>negative_prompt_embeds</code> are generated from the <code>negative_prompt</code> input argument.`,name:"negative_prompt_embeds"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.output_type",description:`<strong>output_type</strong> (<code>str</code>, <em>optional</em>, defaults to <code>&quot;np&quot;</code>) &#x2014;
The output format of the generated video. Choose between <code>torch.Tensor</code> or <code>np.array</code>.`,name:"output_type"},{anchor:"diffusers.TextToVideoSDPipeline.__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_11783/en/api/pipelines/text_to_video#diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput">TextToVideoSDPipelineOutput</a> instead
of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.callback",description:`<strong>callback</strong> (<code>Callable</code>, <em>optional</em>) &#x2014;
A function that calls every <code>callback_steps</code> steps during inference. The function is called with the
following arguments: <code>callback(step: int, timestep: int, latents: torch.Tensor)</code>.`,name:"callback"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.callback_steps",description:`<strong>callback_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 1) &#x2014;
The frequency at which the <code>callback</code> function is called. If not specified, the callback is called at
every step.`,name:"callback_steps"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.cross_attention_kwargs",description:`<strong>cross_attention_kwargs</strong> (<code>dict</code>, <em>optional</em>) &#x2014;
A kwargs dictionary that if specified is passed along to the <code>AttentionProcessor</code> as defined in
<a href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py" rel="nofollow"><code>self.processor</code></a>.`,name:"cross_attention_kwargs"},{anchor:"diffusers.TextToVideoSDPipeline.__call__.clip_skip",description:`<strong>clip_skip</strong> (<code>int</code>, <em>optional</em>) &#x2014;
Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
the output of the pre-final layer will be used for computing the prompt embeddings.`,name:"clip_skip"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth.py#L449",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


<p>If <code>return_dict</code> is <code>True</code>, <a
  href="/docs/diffusers/pr_11783/en/api/pipelines/text_to_video#diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput"
>TextToVideoSDPipelineOutput</a> is
returned, otherwise a <code>tuple</code> is returned where the first element is a list with the generated frames.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>


<p><a
  href="/docs/diffusers/pr_11783/en/api/pipelines/text_to_video#diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput"
>TextToVideoSDPipelineOutput</a> or <code>tuple</code></p>
`}}),C=new so({props:{anchor:"diffusers.TextToVideoSDPipeline.__call__.example",$$slots:{default:[uo]},$$scope:{ctx:J}}}),he=new Ze({props:{name:"encode_prompt",anchor:"diffusers.TextToVideoSDPipeline.encode_prompt",parameters:[{name:"prompt",val:""},{name:"device",val:""},{name:"num_images_per_prompt",val:""},{name:"do_classifier_free_guidance",val:""},{name:"negative_prompt",val:" = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"lora_scale",val:": typing.Optional[float] = None"},{name:"clip_skip",val:": typing.Optional[int] = None"}],parametersDescription:[{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
prompt to be encoded`,name:"prompt"},{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.device",description:`<strong>device</strong> &#x2014; (<code>torch.device</code>):
torch device`,name:"device"},{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.num_images_per_prompt",description:`<strong>num_images_per_prompt</strong> (<code>int</code>) &#x2014;
number of images that should be generated per prompt`,name:"num_images_per_prompt"},{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.do_classifier_free_guidance",description:`<strong>do_classifier_free_guidance</strong> (<code>bool</code>) &#x2014;
whether to use classifier free guidance or not`,name:"do_classifier_free_guidance"},{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.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. If not defined, one has to pass
<code>negative_prompt_embeds</code> instead. Ignored when not using guidance (i.e., ignored if <code>guidance_scale</code> is
less than <code>1</code>).`,name:"negative_prompt"},{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.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.TextToVideoSDPipeline.encode_prompt.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.TextToVideoSDPipeline.encode_prompt.lora_scale",description:`<strong>lora_scale</strong> (<code>float</code>, <em>optional</em>) &#x2014;
A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.`,name:"lora_scale"},{anchor:"diffusers.TextToVideoSDPipeline.encode_prompt.clip_skip",description:`<strong>clip_skip</strong> (<code>int</code>, <em>optional</em>) &#x2014;
Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
the output of the pre-final layer will be used for computing the prompt embeddings.`,name:"clip_skip"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth.py#L159"}}),_e=new P({props:{title:"VideoToVideoSDPipeline",local:"diffusers.VideoToVideoSDPipeline",headingTag:"h2"}}),ve=new Ze({props:{name:"class diffusers.VideoToVideoSDPipeline",anchor:"diffusers.VideoToVideoSDPipeline",parameters:[{name:"vae",val:": AutoencoderKL"},{name:"text_encoder",val:": CLIPTextModel"},{name:"tokenizer",val:": CLIPTokenizer"},{name:"unet",val:": UNet3DConditionModel"},{name:"scheduler",val:": KarrasDiffusionSchedulers"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth_img2img.py#L105"}}),be=new Ze({props:{name:"__call__",anchor:"diffusers.VideoToVideoSDPipeline.__call__",parameters:[{name:"prompt",val:": typing.Union[str, typing.List[str]] = None"},{name:"video",val:": typing.Union[typing.List[numpy.ndarray], torch.Tensor] = None"},{name:"strength",val:": float = 0.6"},{name:"num_inference_steps",val:": int = 50"},{name:"guidance_scale",val:": float = 15.0"},{name:"negative_prompt",val:": typing.Union[str, typing.List[str], NoneType] = None"},{name:"eta",val:": float = 0.0"},{name:"generator",val:": typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None"},{name:"latents",val:": typing.Optional[torch.Tensor] = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"output_type",val:": typing.Optional[str] = 'np'"},{name:"return_dict",val:": bool = True"},{name:"callback",val:": typing.Optional[typing.Callable[[int, int, torch.Tensor], NoneType]] = None"},{name:"callback_steps",val:": int = 1"},{name:"cross_attention_kwargs",val:": typing.Optional[typing.Dict[str, typing.Any]] = None"},{name:"clip_skip",val:": typing.Optional[int] = None"}],parametersDescription:[{anchor:"diffusers.VideoToVideoSDPipeline.__call__.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts to guide image generation. If not defined, you need to pass <code>prompt_embeds</code>.`,name:"prompt"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.video",description:`<strong>video</strong> (<code>List[np.ndarray]</code> or <code>torch.Tensor</code>) &#x2014;
<code>video</code> frames or tensor representing a video batch to be used as the starting point for the process.
Can also accept video latents as <code>image</code>, if passing latents directly, it will not be encoded again.`,name:"video"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.strength",description:`<strong>strength</strong> (<code>float</code>, <em>optional</em>, defaults to 0.8) &#x2014;
Indicates extent to transform the reference <code>video</code>. Must be between 0 and 1. <code>video</code> is used as a
starting point, adding more noise to it the larger the <code>strength</code>. The number of denoising steps
depends on the amount of noise initially added. When <code>strength</code> is 1, added noise is maximum and the
denoising process runs for the full number of iterations specified in <code>num_inference_steps</code>. A value of
1 essentially ignores <code>video</code>.`,name:"strength"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.num_inference_steps",description:`<strong>num_inference_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 50) &#x2014;
The number of denoising steps. More denoising steps usually lead to a higher quality videos at the
expense of slower inference.`,name:"num_inference_steps"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.guidance_scale",description:`<strong>guidance_scale</strong> (<code>float</code>, <em>optional</em>, defaults to 7.5) &#x2014;
A higher guidance scale value encourages the model to generate images closely linked to the text
<code>prompt</code> at the expense of lower image quality. Guidance scale is enabled when <code>guidance_scale &gt; 1</code>.`,name:"guidance_scale"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.negative_prompt",description:`<strong>negative_prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
The prompt or prompts to guide what to not include in video generation. If not defined, you need to
pass <code>negative_prompt_embeds</code> instead. Ignored when not using guidance (<code>guidance_scale &lt; 1</code>).`,name:"negative_prompt"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.eta",description:`<strong>eta</strong> (<code>float</code>, <em>optional</em>, defaults to 0.0) &#x2014;
Corresponds to parameter eta (&#x3B7;) from the <a href="https://huggingface.co/papers/2010.02502" rel="nofollow">DDIM</a> paper. Only
applies to the <a href="/docs/diffusers/pr_11783/en/api/schedulers/ddim#diffusers.DDIMScheduler">DDIMScheduler</a>, and is ignored in other schedulers.`,name:"eta"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.generator",description:`<strong>generator</strong> (<code>torch.Generator</code> or <code>List[torch.Generator]</code>, <em>optional</em>) &#x2014;
A <a href="https://pytorch.org/docs/stable/generated/torch.Generator.html" rel="nofollow"><code>torch.Generator</code></a> to make
generation deterministic.`,name:"generator"},{anchor:"diffusers.VideoToVideoSDPipeline.__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 video
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor is generated by sampling using the supplied random <code>generator</code>. Latents should be of shape
<code>(batch_size, num_channel, num_frames, height, width)</code>.`,name:"latents"},{anchor:"diffusers.VideoToVideoSDPipeline.__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 (prompt weighting). If not
provided, text embeddings are generated from the <code>prompt</code> input argument.`,name:"prompt_embeds"},{anchor:"diffusers.VideoToVideoSDPipeline.__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 (prompt weighting). If
not provided, <code>negative_prompt_embeds</code> are generated from the <code>negative_prompt</code> input argument.`,name:"negative_prompt_embeds"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.output_type",description:`<strong>output_type</strong> (<code>str</code>, <em>optional</em>, defaults to <code>&quot;np&quot;</code>) &#x2014;
The output format of the generated video. Choose between <code>torch.Tensor</code> or <code>np.array</code>.`,name:"output_type"},{anchor:"diffusers.VideoToVideoSDPipeline.__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_11783/en/api/pipelines/text_to_video#diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput">TextToVideoSDPipelineOutput</a> instead
of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.callback",description:`<strong>callback</strong> (<code>Callable</code>, <em>optional</em>) &#x2014;
A function that calls every <code>callback_steps</code> steps during inference. The function is called with the
following arguments: <code>callback(step: int, timestep: int, latents: torch.Tensor)</code>.`,name:"callback"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.callback_steps",description:`<strong>callback_steps</strong> (<code>int</code>, <em>optional</em>, defaults to 1) &#x2014;
The frequency at which the <code>callback</code> function is called. If not specified, the callback is called at
every step.`,name:"callback_steps"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.cross_attention_kwargs",description:`<strong>cross_attention_kwargs</strong> (<code>dict</code>, <em>optional</em>) &#x2014;
A kwargs dictionary that if specified is passed along to the <code>AttentionProcessor</code> as defined in
<a href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py" rel="nofollow"><code>self.processor</code></a>.`,name:"cross_attention_kwargs"},{anchor:"diffusers.VideoToVideoSDPipeline.__call__.clip_skip",description:`<strong>clip_skip</strong> (<code>int</code>, <em>optional</em>) &#x2014;
Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
the output of the pre-final layer will be used for computing the prompt embeddings.`,name:"clip_skip"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth_img2img.py#L514",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


<p>If <code>return_dict</code> is <code>True</code>, <a
  href="/docs/diffusers/pr_11783/en/api/pipelines/text_to_video#diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput"
>TextToVideoSDPipelineOutput</a> is
returned, otherwise a <code>tuple</code> is returned where the first element is a list with the generated frames.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>


<p><a
  href="/docs/diffusers/pr_11783/en/api/pipelines/text_to_video#diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput"
>TextToVideoSDPipelineOutput</a> or <code>tuple</code></p>
`}}),B=new so({props:{anchor:"diffusers.VideoToVideoSDPipeline.__call__.example",$$slots:{default:[go]},$$scope:{ctx:J}}}),Te=new Ze({props:{name:"encode_prompt",anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt",parameters:[{name:"prompt",val:""},{name:"device",val:""},{name:"num_images_per_prompt",val:""},{name:"do_classifier_free_guidance",val:""},{name:"negative_prompt",val:" = None"},{name:"prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"negative_prompt_embeds",val:": typing.Optional[torch.Tensor] = None"},{name:"lora_scale",val:": typing.Optional[float] = None"},{name:"clip_skip",val:": typing.Optional[int] = None"}],parametersDescription:[{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.prompt",description:`<strong>prompt</strong> (<code>str</code> or <code>List[str]</code>, <em>optional</em>) &#x2014;
prompt to be encoded`,name:"prompt"},{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.device",description:`<strong>device</strong> &#x2014; (<code>torch.device</code>):
torch device`,name:"device"},{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.num_images_per_prompt",description:`<strong>num_images_per_prompt</strong> (<code>int</code>) &#x2014;
number of images that should be generated per prompt`,name:"num_images_per_prompt"},{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.do_classifier_free_guidance",description:`<strong>do_classifier_free_guidance</strong> (<code>bool</code>) &#x2014;
whether to use classifier free guidance or not`,name:"do_classifier_free_guidance"},{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.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. If not defined, one has to pass
<code>negative_prompt_embeds</code> instead. Ignored when not using guidance (i.e., ignored if <code>guidance_scale</code> is
less than <code>1</code>).`,name:"negative_prompt"},{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.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.VideoToVideoSDPipeline.encode_prompt.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.VideoToVideoSDPipeline.encode_prompt.lora_scale",description:`<strong>lora_scale</strong> (<code>float</code>, <em>optional</em>) &#x2014;
A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.`,name:"lora_scale"},{anchor:"diffusers.VideoToVideoSDPipeline.encode_prompt.clip_skip",description:`<strong>clip_skip</strong> (<code>int</code>, <em>optional</em>) &#x2014;
Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
the output of the pre-final layer will be used for computing the prompt embeddings.`,name:"clip_skip"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth_img2img.py#L194"}}),ye=new P({props:{title:"TextToVideoSDPipelineOutput",local:"diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput",headingTag:"h2"}}),we=new Ze({props:{name:"class diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput",anchor:"diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput",parameters:[{name:"frames",val:": typing.Union[torch.Tensor, numpy.ndarray, typing.List[typing.List[PIL.Image.Image]]]"}],parametersDescription:[{anchor:"diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput.frames",description:`<strong>frames</strong> (<code>torch.Tensor</code>, <code>np.ndarray</code>, or List[List[PIL.Image.Image]]) &#x2014;
List of video outputs - It can be a nested list of length <code>batch_size,</code> with each sub-list containing
denoised`,name:"frames"}],source:"https://github.com/huggingface/diffusers/blob/vr_11783/src/diffusers/pipelines/text_to_video_synthesis/pipeline_output.py#L13"}}),xe=new co({props:{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/pipelines/text_to_video.md"}}),{c(){a=r("meta"),y=s(),c=r("p"),v=s(),m(T.$$.fragment),p=s(),m(w.$$.fragment),Ce=s(),I=r("div"),I.innerHTML=Wt,De=s(),R=r("p"),R.innerHTML=It,Be=s(),L=r("p"),L.textContent=St,Xe=s(),z=r("p"),z.innerHTML=Ct,Pe=s(),N=r("p"),N.innerHTML=Dt,Re=s(),m(E.$$.fragment),Le=s(),m(Y.$$.fragment),ze=s(),H=r("p"),H.textContent=Bt,Ne=s(),m(F.$$.fragment),Ee=s(),Q=r("p"),Q.textContent=Xt,Ye=s(),q=r("p"),q.textContent=Pt,He=s(),m(A.$$.fragment),Fe=s(),O=r("p"),O.innerHTML=Rt,Qe=s(),K=r("p"),K.textContent=Lt,qe=s(),m(ee.$$.fragment),Ae=s(),te=r("p"),te.textContent=zt,Oe=s(),oe=r("table"),oe.innerHTML=Nt,Ke=s(),m(ne.$$.fragment),et=s(),se=r("p"),se.innerHTML=Et,tt=s(),m(ie.$$.fragment),ot=s(),ae=r("p"),ae.textContent=Yt,nt=s(),m(re.$$.fragment),st=s(),le=r("p"),le.textContent=Ht,it=s(),pe=r("table"),pe.innerHTML=Ft,at=s(),m(de.$$.fragment),rt=s(),ce=r("p"),ce.innerHTML=Qt,lt=s(),me=r("p"),me.innerHTML=qt,pt=s(),m(S.$$.fragment),dt=s(),m(fe.$$.fragment),ct=s(),x=r("div"),m(ue.$$.fragment),yt=s(),V=r("div"),m(ge.$$.fragment),wt=s(),Ve=r("p"),Ve.textContent=At,xt=s(),m(C.$$.fragment),Mt=s(),D=r("div"),m(he.$$.fragment),Zt=s(),je=r("p"),je.textContent=Ot,mt=s(),m(_e.$$.fragment),ft=s(),M=r("div"),m(ve.$$.fragment),Jt=s(),j=r("div"),m(be.$$.fragment),Vt=s(),$e=r("p"),$e.textContent=Kt,jt=s(),m(B.$$.fragment),$t=s(),X=r("div"),m(Te.$$.fragment),kt=s(),ke=r("p"),ke.textContent=eo,ut=s(),m(ye.$$.fragment),gt=s(),Z=r("div"),m(we.$$.fragment),Gt=s(),Ge=r("p"),Ge.textContent=to,Ut=s(),Ue=r("p"),Ue.innerHTML=oo,ht=s(),m(xe.$$.fragment),_t=s(),Se=r("p"),this.h()},l(e){const t=po("svelte-u9bgzb",document.head);a=l(t,"META",{name:!0,content:!0}),t.forEach(o),y=i(e),c=l(e,"P",{}),W(c).forEach(o),v=i(e),f(T.$$.fragment,e),p=i(e),f(w.$$.fragment,e),Ce=i(e),I=l(e,"DIV",{class:!0,"data-svelte-h":!0}),d(I)!=="svelte-si9ct8"&&(I.innerHTML=Wt),De=i(e),R=l(e,"P",{"data-svelte-h":!0}),d(R)!=="svelte-nuyszd"&&(R.innerHTML=It),Be=i(e),L=l(e,"P",{"data-svelte-h":!0}),d(L)!=="svelte-1cwsb16"&&(L.textContent=St),Xe=i(e),z=l(e,"P",{"data-svelte-h":!0}),d(z)!=="svelte-1ovx0um"&&(z.innerHTML=Ct),Pe=i(e),N=l(e,"P",{"data-svelte-h":!0}),d(N)!=="svelte-9292p9"&&(N.innerHTML=Dt),Re=i(e),f(E.$$.fragment,e),Le=i(e),f(Y.$$.fragment,e),ze=i(e),H=l(e,"P",{"data-svelte-h":!0}),d(H)!=="svelte-a4ktac"&&(H.textContent=Bt),Ne=i(e),f(F.$$.fragment,e),Ee=i(e),Q=l(e,"P",{"data-svelte-h":!0}),d(Q)!=="svelte-1gwbshq"&&(Q.textContent=Xt),Ye=i(e),q=l(e,"P",{"data-svelte-h":!0}),d(q)!=="svelte-yoc5w4"&&(q.textContent=Pt),He=i(e),f(A.$$.fragment,e),Fe=i(e),O=l(e,"P",{"data-svelte-h":!0}),d(O)!=="svelte-m9rmsr"&&(O.innerHTML=Rt),Qe=i(e),K=l(e,"P",{"data-svelte-h":!0}),d(K)!=="svelte-1yxs7ur"&&(K.textContent=Lt),qe=i(e),f(ee.$$.fragment,e),Ae=i(e),te=l(e,"P",{"data-svelte-h":!0}),d(te)!=="svelte-1pxzje"&&(te.textContent=zt),Oe=i(e),oe=l(e,"TABLE",{"data-svelte-h":!0}),d(oe)!=="svelte-5s26g4"&&(oe.innerHTML=Nt),Ke=i(e),f(ne.$$.fragment,e),et=i(e),se=l(e,"P",{"data-svelte-h":!0}),d(se)!=="svelte-gtl6zm"&&(se.innerHTML=Et),tt=i(e),f(ie.$$.fragment,e),ot=i(e),ae=l(e,"P",{"data-svelte-h":!0}),d(ae)!=="svelte-4ywac8"&&(ae.textContent=Yt),nt=i(e),f(re.$$.fragment,e),st=i(e),le=l(e,"P",{"data-svelte-h":!0}),d(le)!=="svelte-1pxzje"&&(le.textContent=Ht),it=i(e),pe=l(e,"TABLE",{"data-svelte-h":!0}),d(pe)!=="svelte-9iw63g"&&(pe.innerHTML=Ft),at=i(e),f(de.$$.fragment,e),rt=i(e),ce=l(e,"P",{"data-svelte-h":!0}),d(ce)!=="svelte-op8wgu"&&(ce.innerHTML=Qt),lt=i(e),me=l(e,"P",{"data-svelte-h":!0}),d(me)!=="svelte-1lb9rti"&&(me.innerHTML=qt),pt=i(e),f(S.$$.fragment,e),dt=i(e),f(fe.$$.fragment,e),ct=i(e),x=l(e,"DIV",{class:!0});var k=W(x);f(ue.$$.fragment,k),yt=i(k),V=l(k,"DIV",{class:!0});var G=W(V);f(ge.$$.fragment,G),wt=i(G),Ve=l(G,"P",{"data-svelte-h":!0}),d(Ve)!=="svelte-50j04k"&&(Ve.textContent=At),xt=i(G),f(C.$$.fragment,G),G.forEach(o),Mt=i(k),D=l(k,"DIV",{class:!0});var Me=W(D);f(he.$$.fragment,Me),Zt=i(Me),je=l(Me,"P",{"data-svelte-h":!0}),d(je)!=="svelte-16q0ax1"&&(je.textContent=Ot),Me.forEach(o),k.forEach(o),mt=i(e),f(_e.$$.fragment,e),ft=i(e),M=l(e,"DIV",{class:!0});var U=W(M);f(ve.$$.fragment,U),Jt=i(U),j=l(U,"DIV",{class:!0});var We=W(j);f(be.$$.fragment,We),Vt=i(We),$e=l(We,"P",{"data-svelte-h":!0}),d($e)!=="svelte-50j04k"&&($e.textContent=Kt),jt=i(We),f(B.$$.fragment,We),We.forEach(o),$t=i(U),X=l(U,"DIV",{class:!0});var bt=W(X);f(Te.$$.fragment,bt),kt=i(bt),ke=l(bt,"P",{"data-svelte-h":!0}),d(ke)!=="svelte-16q0ax1"&&(ke.textContent=eo),bt.forEach(o),U.forEach(o),ut=i(e),f(ye.$$.fragment,e),gt=i(e),Z=l(e,"DIV",{class:!0});var Ie=W(Z);f(we.$$.fragment,Ie),Gt=i(Ie),Ge=l(Ie,"P",{"data-svelte-h":!0}),d(Ge)!=="svelte-1uuckb2"&&(Ge.textContent=to),Ut=i(Ie),Ue=l(Ie,"P",{"data-svelte-h":!0}),d(Ue)!=="svelte-gk6g69"&&(Ue.innerHTML=oo),Ie.forEach(o),ht=i(e),f(xe.$$.fragment,e),_t=i(e),Se=l(e,"P",{}),W(Se).forEach(o),this.h()},h(){$(a,"name","hf:doc:metadata"),$(a,"content",_o),$(I,"class","flex flex-wrap space-x-1"),$(V,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(D,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(x,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(j,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(X,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(M,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),$(Z,"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){b(document.head,a),n(e,y,t),n(e,c,t),n(e,v,t),u(T,e,t),n(e,p,t),u(w,e,t),n(e,Ce,t),n(e,I,t),n(e,De,t),n(e,R,t),n(e,Be,t),n(e,L,t),n(e,Xe,t),n(e,z,t),n(e,Pe,t),n(e,N,t),n(e,Re,t),u(E,e,t),n(e,Le,t),u(Y,e,t),n(e,ze,t),n(e,H,t),n(e,Ne,t),u(F,e,t),n(e,Ee,t),n(e,Q,t),n(e,Ye,t),n(e,q,t),n(e,He,t),u(A,e,t),n(e,Fe,t),n(e,O,t),n(e,Qe,t),n(e,K,t),n(e,qe,t),u(ee,e,t),n(e,Ae,t),n(e,te,t),n(e,Oe,t),n(e,oe,t),n(e,Ke,t),u(ne,e,t),n(e,et,t),n(e,se,t),n(e,tt,t),u(ie,e,t),n(e,ot,t),n(e,ae,t),n(e,nt,t),u(re,e,t),n(e,st,t),n(e,le,t),n(e,it,t),n(e,pe,t),n(e,at,t),u(de,e,t),n(e,rt,t),n(e,ce,t),n(e,lt,t),n(e,me,t),n(e,pt,t),u(S,e,t),n(e,dt,t),u(fe,e,t),n(e,ct,t),n(e,x,t),u(ue,x,null),b(x,yt),b(x,V),u(ge,V,null),b(V,wt),b(V,Ve),b(V,xt),u(C,V,null),b(x,Mt),b(x,D),u(he,D,null),b(D,Zt),b(D,je),n(e,mt,t),u(_e,e,t),n(e,ft,t),n(e,M,t),u(ve,M,null),b(M,Jt),b(M,j),u(be,j,null),b(j,Vt),b(j,$e),b(j,jt),u(B,j,null),b(M,$t),b(M,X),u(Te,X,null),b(X,kt),b(X,ke),n(e,ut,t),u(ye,e,t),n(e,gt,t),n(e,Z,t),u(we,Z,null),b(Z,Gt),b(Z,Ge),b(Z,Ut),b(Z,Ue),n(e,ht,t),u(xe,e,t),n(e,_t,t),n(e,Se,t),vt=!0},p(e,[t]){const k={};t&2&&(k.$$scope={dirty:t,ctx:e}),T.$set(k);const G={};t&2&&(G.$$scope={dirty:t,ctx:e}),S.$set(G);const Me={};t&2&&(Me.$$scope={dirty:t,ctx:e}),C.$set(Me);const U={};t&2&&(U.$$scope={dirty:t,ctx:e}),B.$set(U)},i(e){vt||(g(T.$$.fragment,e),g(w.$$.fragment,e),g(E.$$.fragment,e),g(Y.$$.fragment,e),g(F.$$.fragment,e),g(A.$$.fragment,e),g(ee.$$.fragment,e),g(ne.$$.fragment,e),g(ie.$$.fragment,e),g(re.$$.fragment,e),g(de.$$.fragment,e),g(S.$$.fragment,e),g(fe.$$.fragment,e),g(ue.$$.fragment,e),g(ge.$$.fragment,e),g(C.$$.fragment,e),g(he.$$.fragment,e),g(_e.$$.fragment,e),g(ve.$$.fragment,e),g(be.$$.fragment,e),g(B.$$.fragment,e),g(Te.$$.fragment,e),g(ye.$$.fragment,e),g(we.$$.fragment,e),g(xe.$$.fragment,e),vt=!0)},o(e){h(T.$$.fragment,e),h(w.$$.fragment,e),h(E.$$.fragment,e),h(Y.$$.fragment,e),h(F.$$.fragment,e),h(A.$$.fragment,e),h(ee.$$.fragment,e),h(ne.$$.fragment,e),h(ie.$$.fragment,e),h(re.$$.fragment,e),h(de.$$.fragment,e),h(S.$$.fragment,e),h(fe.$$.fragment,e),h(ue.$$.fragment,e),h(ge.$$.fragment,e),h(C.$$.fragment,e),h(he.$$.fragment,e),h(_e.$$.fragment,e),h(ve.$$.fragment,e),h(be.$$.fragment,e),h(B.$$.fragment,e),h(Te.$$.fragment,e),h(ye.$$.fragment,e),h(we.$$.fragment,e),h(xe.$$.fragment,e),vt=!1},d(e){e&&(o(y),o(c),o(v),o(p),o(Ce),o(I),o(De),o(R),o(Be),o(L),o(Xe),o(z),o(Pe),o(N),o(Re),o(Le),o(ze),o(H),o(Ne),o(Ee),o(Q),o(Ye),o(q),o(He),o(Fe),o(O),o(Qe),o(K),o(qe),o(Ae),o(te),o(Oe),o(oe),o(Ke),o(et),o(se),o(tt),o(ot),o(ae),o(nt),o(st),o(le),o(it),o(pe),o(at),o(rt),o(ce),o(lt),o(me),o(pt),o(dt),o(ct),o(x),o(mt),o(ft),o(M),o(ut),o(gt),o(Z),o(ht),o(_t),o(Se)),o(a),_(T,e),_(w,e),_(E,e),_(Y,e),_(F,e),_(A,e),_(ee,e),_(ne,e),_(ie,e),_(re,e),_(de,e),_(S,e),_(fe,e),_(ue),_(ge),_(C),_(he),_(_e,e),_(ve),_(be),_(B),_(Te),_(ye,e),_(we),_(xe,e)}}}const _o='{"title":"Text-to-video","local":"text-to-video","sections":[{"title":"Usage example","local":"usage-example","sections":[{"title":"text-to-video-ms-1.7b","local":"text-to-video-ms-17b","sections":[],"depth":3},{"title":"cerspense/zeroscope_v2_576w & cerspense/zeroscope_v2_XL","local":"cerspensezeroscopev2576w--cerspensezeroscopev2xl","sections":[],"depth":3}],"depth":2},{"title":"Tips","local":"tips","sections":[],"depth":2},{"title":"TextToVideoSDPipeline","local":"diffusers.TextToVideoSDPipeline","sections":[],"depth":2},{"title":"VideoToVideoSDPipeline","local":"diffusers.VideoToVideoSDPipeline","sections":[],"depth":2},{"title":"TextToVideoSDPipelineOutput","local":"diffusers.pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput","sections":[],"depth":2}],"depth":1}';function vo(J){return ao(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class Jo extends ro{constructor(a){super(),lo(this,a,vo,ho,io,{})}}export{Jo as component};