Buckets:

hf-doc-build
/

doc

Files
xet
hf-doc-build/doc / diffusers /main /ko /api /pipelines /minimax_h3_image.html
mishig
1 day ago
download
raw
43.8 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;MiniMax-H3-Image&quot;,&quot;local&quot;:&quot;minimax-h3-image&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;구성 요소&quot;,&quot;local&quot;:&quot;components&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;워크플로&quot;,&quot;local&quot;:&quot;workflows&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;프롬프트 향상&quot;,&quot;local&quot;:&quot;prompt-enhancement&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;이미지 크기&quot;,&quot;local&quot;:&quot;image-size&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;샘플링&quot;,&quot;local&quot;:&quot;sampling&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;MiniMaxH3ImageModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.MiniMaxH3ImageModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;MiniMaxH3ImageBlocks&quot;,&quot;local&quot;:&quot;diffusers.MiniMaxH3ImageBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
		<link href="/docs/diffusers/main/en/_app/immutable/entry/start.U6TUchwJ.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/yxSHiGDF.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/CPfOwUO_.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/entry/app.3qKhPoX-.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/BYe-f74j.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/DnVIJfC5.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/D32hXVzh.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/0.DqpuBF3n.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Dyp9xKq8.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/9.BHf5c0WW.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Dk44y1gU.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/C4N0BX8L.js" rel="modulepreload">
		<!--mp1ecb--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;MiniMax-H3-Image&quot;,&quot;local&quot;:&quot;minimax-h3-image&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Components&quot;,&quot;local&quot;:&quot;components&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Workflows&quot;,&quot;local&quot;:&quot;workflows&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Prompt enhancement&quot;,&quot;local&quot;:&quot;prompt-enhancement&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Image size&quot;,&quot;local&quot;:&quot;image-size&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Sampling&quot;,&quot;local&quot;:&quot;sampling&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;MiniMaxH3ImageModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.MiniMaxH3ImageModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;MiniMaxH3ImageBlocks&quot;,&quot;local&quot;:&quot;diffusers.MiniMaxH3ImageBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/><!---->
		<link href="/docs/diffusers/main/en/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: inline-end; margin-inline-start: 10px; display: inline-flex; position: relative; z-index: 10;"><div dir="ltr" class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----></span> <span dir="auto">Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg><!----></button></div> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="minimax-h3-image" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#minimax-h3-image"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->MiniMax-H3-Image<!----></span></h1><!--]--><!----> <p>MiniMax-H3-Image는 <a href="./minimax_h3">MiniMax-H3</a>를 기반으로 한 MiniMax의 이미지 모델로, 텍스트 기반 이미지 생성 및 참조 이미지 편집에 사용됩니다. 체크포인트는 <a href="https://huggingface.co/MiniMaxAI/MiniMax-H3-Image" rel="nofollow">MiniMaxAI/MiniMax-H3-Image</a>에서 찾을 수 있습니다.</p> <p>MiniMax-H3와 마찬가지로, <a href="../../modular_diffusers/overview">Modular Diffusers</a> 블록으로만 통합됩니다: <a href="/docs/diffusers/main/ko/api/pipelines/minimax_h3_image#diffusers.MiniMaxH3ImageBlocks">MiniMaxH3ImageBlocks</a>와 해당 <a href="/docs/diffusers/main/ko/api/pipelines/minimax_h3_image#diffusers.MiniMaxH3ImageModularPipeline">MiniMaxH3ImageModularPipeline</a>이 전체 통합을 구성합니다.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> 복사됨</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> ComponentsManager, ModularPipeline

manager = ComponentsManager()
manager.enable_auto_cpu_offload(device=<span class="hljs-string">&quot;cuda&quot;</span>)

pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;MiniMaxAI/MiniMax-H3-Image&quot;</span>, components_manager=manager)
pipe.load_components(dtype=torch.bfloat16)<!----></pre></div><!----> <p>transformer와 text encoder는 bfloat16에서 각각 약 62 GB로, 80 GB GPU에 함께 올라가지 않으므로 예제에서는 auto offloading을 사용합니다. <a href="/docs/diffusers/main/ko/api/modular_diffusers/pipeline_components#diffusers.ComponentsManager">ComponentsManager</a>는 블록이 해당 구성 요소를 필요로 할 때 이를 accelerator로 이동합니다. 4736x3552 이미지의 경우 피크 메모리가 약 65 GB까지 올라갑니다.</p> <p>텍스트-이미지:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> 복사됨</div><!----></button><!----></div> <pre class="language-py "><!---->prompt = (
    <span class="hljs-string">&quot;An overhead photograph of a rustic farm-to-table spread on a weathered wooden table: a round loaf of sourdough &quot;</span>
    <span class="hljs-string">&quot;on burlap, wedges of cheese on a dark board, a copper bowl of vegetables, vine tomatoes, olives and walnuts.&quot;</span>
)
image = pipe(
    prompt=prompt, height=<span class="hljs-number">3552</span>, width=<span class="hljs-number">4736</span>, generator=torch.Generator().manual_seed(<span class="hljs-number">42</span>), output=<span class="hljs-string">&quot;images&quot;</span>
)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <p>참조 이미지 편집:</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> 복사됨</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

reference = load_image(<span class="hljs-string">&quot;https://huggingface.co/MiniMaxAI/MiniMax-H3-Image/resolve/main/examples/reference1.png&quot;</span>)
image = pipe(
    prompt=<span class="hljs-string">&quot;Keep everything in &lt;Picture 1&gt; as it is, but remove the headphones and have the two hosts fist bump.&quot;</span>,
    image=reference,
    height=<span class="hljs-number">4096</span>,
    width=<span class="hljs-number">4096</span>,
    generator=torch.Generator().manual_seed(<span class="hljs-number">1234</span>),
    output=<span class="hljs-string">&quot;images&quot;</span>,
)[<span class="hljs-number">0</span>]<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="components" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#components"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->구성 요소<!----></span></h2><!--]--><!----> <table><thead><tr><th>컴포넌트</th><th>클래스</th></tr></thead><tbody><tr><td><code>transformer</code></td><td><a href="/docs/diffusers/main/ko/api/models/minimax_h3_image_transformer2d#diffusers.MiniMaxH3ImageTransformer2DModel">MiniMaxH3ImageTransformer2DModel</a></td></tr><tr><td><code>text_encoder</code></td><td><code>Qwen3VLForConditionalGeneration</code></td></tr><tr><td><code>vae</code></td><td><a href="/docs/diffusers/main/ko/api/models/autoencoderkl_minimax_h3_image#diffusers.AutoencoderKLMiniMaxH3Image">AutoencoderKLMiniMaxH3Image</a></td></tr><tr><td><code>tokenizer</code>, <code>processor</code></td><td>텍스트 인코더의 토크나이저와 <code>Qwen3VLProcessor</code></td></tr><tr><td><code>scheduler</code></td><td><a href="/docs/diffusers/main/ko/api/schedulers/minimax_h3#diffusers.MiniMaxH3Scheduler">MiniMaxH3Scheduler</a></td></tr></tbody></table> <p>텍스트 인코더는 <a href="./minimax_h3">MiniMax-H3</a>에서 사용하는 것과 동일합니다. 저장소가 자체적으로 완결되도록 체크포인트에는 해당 인코더의 사본이 포함됩니다. Hub는 저장소 간에 동일한 파일을 중복 제거하므로, 이 사본은 Hub에서 추가 저장 공간을 차지하지 않습니다.</p> <!--[1--><h2 class="relative group"><a id="workflows" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#workflows"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->워크플로<!----></span></h2><!--]--><!----> <p>워크플로우는 <code>image</code> 입력에서 선택됩니다:</p> <ul><li><strong><code>text2image</code></strong>는 <code>prompt</code>로부터 생성합니다.</li> <li><strong><code>image_conditioned</code></strong>는 <code>prompt</code>와 <code>image</code>로 전달된 하나 이상의 참조 이미지로부터 생성합니다. 프롬프트는 참조 이미지를 순서대로 <code>&lt;Picture 1></code>, <code>&lt;Picture 2></code> 등으로 지칭합니다. 참조 이미지는 자체 크기를 유지하며, 짧은 변의 길이가 1024~2048픽셀 범위를 벗어난 경우에만 리사이즈되고, 생성된 이미지의 크기로 리사이즈되지 않습니다.</li></ul> <p>프롬프트는 채팅 템플릿 없이 그대로 사용되며, 길고 완전한 설명이 가장 효과적입니다. 편집 프롬프트는 일반적으로 유지할 요소를 명시한 후 변경할 요소를 명시하고, 마지막으로 대상 이미지를 완전히 설명하는 구조를 가집니다.</p> <!--[1--><h2 class="relative group"><a id="prompt-enhancement" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#prompt-enhancement"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->프롬프트 향상<!----></span></h2><!--]--><!----> <p><a href="https://huggingface.co/MiniMaxAI/MiniMax-H3-Image-PE" rel="nofollow">MiniMax-H3-Image-PE</a>는 MiniMax의 프롬프트 확장 모델입니다. 짧은 요청을 이미지 모델이 가장 잘 처리할 수 있는 수준의 상세한 설명으로 재작성하고, 권장 종횡비를 제안합니다. <code>diffusers/minimax-h3-image-pe-blocks</code> 커스텀 블록은 이를 자체 파이프라인으로 실행합니다. <code>xgrammar</code>가 필요하며, GPU 메모리 약 18 GB가 필요합니다.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> 복사됨</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> ModularPipeline

enhancer = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;diffusers/minimax-h3-image-pe-blocks&quot;</span>, trust_remote_code=<span class="hljs-literal">True</span>)
enhancer.load_components(dtype=torch.bfloat16, device_map=<span class="hljs-string">&quot;cuda&quot;</span>)

state = enhancer(prompt=<span class="hljs-string">&quot;a cozy cabin in the snow&quot;</span>)
prompt, aspect_ratio = state.get(<span class="hljs-string">&quot;prompt&quot;</span>), state.get(<span class="hljs-string">&quot;aspect_ratio&quot;</span>)<!----></pre></div><!----> <p>편집 지시문을 확장하려면 참조 이미지를 <code>image</code>로 전달하세요. 그런 다음 <code>aspect_ratio</code>와 일치하는 <code>height</code>와 <code>width</code>를 지정하여 <code>prompt</code>를 이미지 파이프라인에 전달하세요.</p> <!--[1--><h2 class="relative group"><a id="image-size" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#image-size"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->이미지 크기<!----></span></h2><!--]--><!----> <p><code>height</code>와 <code>width</code>는 반환되는 이미지의 크기를 나타내며 기본값은 2048입니다. 모델은 4 메가픽셀 이상으로 생성합니다. 요청된 크기가 더 작으면 동일한 종횡비인 더 큰 캔버스에서 생성한 뒤 원래 크기로 다시 조정되므로, 1024x1024 요청은 2048x2048 요청과 동일한 비용이 발생합니다. <code>min_generation_area</code>는 픽셀 단위로 해당 임계값을 설정합니다. 요청한 크기로 항상 생성하려면 <code>1024</code>를 전달하세요. 어느 경우든 캔버스의 각 변은 32의 배수로 반올림됩니다.</p> <p>MiniMax는 1:1 비율에 4096x4096, 4:3 비율에 4736x3552, 3:2 비율에 5024x3360, 16:9 비율에 5472x3072, 21:9 비율에 6272x2688을 권장하며, 세로 형식의 경우 동일한 크기의 높이와 너비를 서로 바꾸어 사용하도록 권장합니다.</p> <!--[1--><h2 class="relative group"><a id="sampling" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#sampling"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->샘플링<!----></span></h2><!--]--><!----> <ul><li>체크포인트는 <code>sample_sigmas</code>에 대해 6단계로 샘플링하며, 트랜스포머에는 단계마다 하나의 출력 헤드가 있으므로 <code>num_inference_steps</code>가 없습니다.</li> <li>모델은 guidance-distilled이므로 guider, <code>negative_prompt</code> 또는 <code>guidance_scale</code>이 없습니다.</li></ul> <!--[1--><h2 class="relative group"><a id="diffusers.MiniMaxH3ImageModularPipeline" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#diffusers.MiniMaxH3ImageModularPipeline"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->MiniMaxH3ImageModularPipeline<!----></span></h2><!--]--><!----> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><div><span class="group flex space-x-1.5 items-center text-gray-800 bg-gradient-to-r rounded-tr-lg -mt-4 -ml-4 pt-3 px-2.5" id="diffusers.MiniMaxH3ImageModularPipeline"><!----><h3 class="!m-0"><span class="flex-1 break-all md:text-lg bg-gradient-to-r px-2.5 py-1.5 rounded-xl from-indigo-50/70 to-white dark:from-gray-900 dark:to-gray-950 dark:text-indigo-300 text-indigo-700"><svg class="mr-1.5 text-indigo-500 inline-block -mt-0.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width=".8em" height=".8em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 24 24"><path class="uim-quaternary" d="M20.23 7.24L12 12L3.77 7.24a1.98 1.98 0 0 1 .7-.71L11 2.76c.62-.35 1.38-.35 2 0l6.53 3.77c.29.173.531.418.7.71z" opacity=".25" fill="currentColor"></path><path class="uim-tertiary" d="M12 12v9.5a2.09 2.09 0 0 1-.91-.21L4.5 17.48a2.003 2.003 0 0 1-1-1.73v-7.5a2.06 2.06 0 0 1 .27-1.01L12 12z" opacity=".5" fill="currentColor"></path><path class="uim-primary" d="M20.5 8.25v7.5a2.003 2.003 0 0 1-1 1.73l-6.62 3.82c-.275.13-.576.198-.88.2V12l8.23-4.76c.175.308.268.656.27 1.01z" fill="currentColor"></path></svg><span class="font-light">class</span> <span class="font-medium">diffusers.</span><span class="font-semibold">MiniMaxH3ImageModularPipeline</span></span></h3><!----> <a id="diffusers.MiniMaxH3ImageModularPipeline" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.MiniMaxH3ImageModularPipeline"><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></a> <!--[0--><a class="!ml-auto !text-gray-400 !no-underline text-sm flex items-center" href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/modular_pipelines/minimax_h3_image/modular_pipeline.py#L27" target="_blank"><span>&lt;</span> <span class="hidden md:block mx-0.5 hover:!underline">source</span> <span>></span></a><!--]--></span> <!--[0--><p class="font-mono text-xs md:text-sm !leading-relaxed !my-6"><span>(</span> <!--[--><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">blocks<span class="opacity-60">: diffusers.modular_pipelines.modular_pipeline.ModularPipelineBlocks | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">pretrained_model_name_or_path<span class="opacity-60">: str | os.PathLike | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">components_manager<span class="opacity-60">: diffusers.modular_pipelines.components_manager.ComponentsManager | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">collection<span class="opacity-60">: str | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">workflow<span class="opacity-60">: str | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">modular_config_dict<span class="opacity-60">: dict[str, typing.Any] | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">config_dict<span class="opacity-60">: dict[str, typing.Any] | None = None</span></span></span><span class="comma cursor-default"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">**kwargs<span class="opacity-60"></span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>MiniMax-H3-Image용 ModularPipeline: 텍스트-이미지 생성(<code>text2image</code>) 및 참조 이미지 편집(<code>image_conditioned</code>)</p> <p>MiniMax-H3-Image는 오디오 없이 텍스트 행, 참조 이미지 행, 생성된 이미지 행(단일 latent 프레임)으로 구성된 하나의 packed sequence 위에서 MiniMax-H3 transformer를 실행합니다. 체크포인트는 두 번 증류됩니다. 가이던스가 가중치에 내장되어 있어 guider, <code>negative_prompt</code>, <code>guidance_scale</code>이 없으며, transformer의 출력 헤드 하나당 한 단계씩 고정된 6단계 그리드에서 샘플링하므로 <code>num_inference_steps</code>도 없습니다.</p> <div class="relative group rounded-md"><a id="diffusers.MiniMaxH3ImageModularPipeline.example" class="header-link block pr-0.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#diffusers.MiniMaxH3ImageModularPipeline.example"><span><svg class="text-smd" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> 복사됨</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> ModularPipeline

pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;MiniMaxAI/MiniMax-H3-Image&quot;</span>)
pipe.load_components(dtype=torch.bfloat16)<!----></pre></div><!----></div><!----></div> <!--[1--><h2 class="relative group"><a id="diffusers.MiniMaxH3ImageBlocks" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#diffusers.MiniMaxH3ImageBlocks"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->MiniMaxH3ImageBlocks<!----></span></h2><!--]--><!----> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><div><span class="group flex space-x-1.5 items-center text-gray-800 bg-gradient-to-r rounded-tr-lg -mt-4 -ml-4 pt-3 px-2.5" id="diffusers.MiniMaxH3ImageBlocks"><!----><h3 class="!m-0"><span class="flex-1 break-all md:text-lg bg-gradient-to-r px-2.5 py-1.5 rounded-xl from-indigo-50/70 to-white dark:from-gray-900 dark:to-gray-950 dark:text-indigo-300 text-indigo-700"><svg class="mr-1.5 text-indigo-500 inline-block -mt-0.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width=".8em" height=".8em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 24 24"><path class="uim-quaternary" d="M20.23 7.24L12 12L3.77 7.24a1.98 1.98 0 0 1 .7-.71L11 2.76c.62-.35 1.38-.35 2 0l6.53 3.77c.29.173.531.418.7.71z" opacity=".25" fill="currentColor"></path><path class="uim-tertiary" d="M12 12v9.5a2.09 2.09 0 0 1-.91-.21L4.5 17.48a2.003 2.003 0 0 1-1-1.73v-7.5a2.06 2.06 0 0 1 .27-1.01L12 12z" opacity=".5" fill="currentColor"></path><path class="uim-primary" d="M20.5 8.25v7.5a2.003 2.003 0 0 1-1 1.73l-6.62 3.82c-.275.13-.576.198-.88.2V12l8.23-4.76c.175.308.268.656.27 1.01z" fill="currentColor"></path></svg><span class="font-light">클래스</span> <span class="font-medium">diffusers.</span><span class="font-semibold">MiniMaxH3ImageBlocks</span></span></h3><!----> <a id="diffusers.MiniMaxH3ImageBlocks" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.MiniMaxH3ImageBlocks"><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></a> <!--[0--><a class="!ml-auto !text-gray-400 !no-underline text-sm flex items-center" href="https://github.com/huggingface/diffusers/blob/main/src/diffusers/modular_pipelines/minimax_h3_image/modular_blocks_minimax_h3_image.py#L371" target="_blank"><span>&lt;</span> <span class="hidden md:block mx-0.5 hover:!underline">source</span> <span>></span></a><!--]--></span> <!--[0--><p class="font-mono text-xs md:text-sm !leading-relaxed !my-6"><span>(</span> <!--[--><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>MiniMax-H3-Image의 모듈형 파이프라인 블록: <code>text2image</code> 및 <code>image</code>가 전달될 때의 참조 이미지 편집(<code>image_conditioned</code>).</p> <p>지원되는 워크플로:</p> <ul><li><code>text2image</code>: <code>prompt</code> 필수</li> <li><code>image_conditioned</code>: <code>prompt</code>, <code>image</code> 필수</li></ul> <p>구성 요소:
image_processor (<code>VaeImageProcessor</code>) text_encoder (<code>Qwen3VLForConditionalGeneration</code>) tokenizer
(<code>Qwen2Tokenizer</code>) processor (<code>Qwen3VLProcessor</code>) vae (<code>AutoencoderKLMiniMaxH3Image</code>) scheduler
(<code>MiniMaxH3Scheduler</code>) transformer (<code>MiniMaxH3ImageTransformer2DModel</code>)</p> <p>Configs:
reference_min_short_edge (기본값: 1024) reference_max_short_edge (기본값: 2048) vision_min_short_edge
(기본값: 1024) vision_max_short_edge (기본값: 1024) sample_sigmas (기본값: [1.0, 0.9675236940383911,
0.8875485062599182, 0.75, 0.5327739119529724, 0.23200887441635132, 0.0]): 체크포인트의 단계별 출력 헤드가 증류(distill)된 샘플링 그리드입니다.</p> <p>입력:
image (<code>Image | list</code>, <em>선택 사항</em>):
편집의 참조 이미지 또는 이미지 목록입니다. 프롬프트는 순서대로 <code>&lt;Picture 1></code>, <code>&lt;Picture 2></code>, …로 참조합니다.
prompt (<code>str</code>):
생성을 안내하는 프롬프트, 단일 문자열입니다.
vision_images (<code>list</code>, <em>선택 사항</em>):
텍스트 인코더를 위해 리사이즈된 참조 이미지입니다.
reference_images (<code>list</code>, <em>선택 사항</em>):
VAE를 위해 리사이즈된 참조 이미지입니다.
generator (<code>Generator</code>, <em>선택 사항</em>):
요청의 생성기입니다. 각 참조마다 하나의 사후 샘플이 여기서 추출됩니다.
height (<code>int</code>):
반환되는 이미지의 픽셀 단위 높이입니다.
width (<code>int</code>):
반환되는 이미지의 픽셀 단위 너비입니다.
min_generation_area (<code>int</code>, <em>선택 사항</em>, 기본값 4194304):
이미지가 생성되는 최소 픽셀 단위 면적입니다. 더 작은 <code>height x width</code>는 이 면적으로 확대된 캔버스에서 생성된 후 디코딩을 거쳐 <code>height x width</code>로 다시 리사이즈됩니다. 요청한 크도로 항상 생성하려면 1024를 전달하세요.
reference_latents (<code>list</code>, <em>선택 사항</em>):
참조 이미지의 조건부 잠재 변수입니다. 각 참조의 기하학적 구조가 여기서 유래합니다.
latents (<code>Tensor</code>):
<code>(1, latent_channels, latent_height, latent_width)</code> 형태의 사전 생성된 노이즈로, 추출 대신 사용됩니다.
reference_noise (<code>list</code>, <em>선택 사항</em>):
참조 이미지의 사전 생성된 노이즈로, <code>reference_latents</code>의 각 항목과 동일한 형태의 텐서 하나씩이며, 추출 대신 사용됩니다.
*<em>denoiser_input_fields (<code>None</code>,</em>선택 사항<em>):
트랜스포머가 이름으로 읽는 패킹된 시퀀스의 구조적 설명: <code>token_tags</code>, <code>position_ids</code> 및 두 행 인덱스 텐서입니다.
output_type (<code>str</code>,</em>선택 사항*, 기본값 pil):
출력 형식: ‘pil’, ‘np’, ‘pt’.</p> <p>출력:
images (<code>list</code>):
생성된 이미지.</p></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/pipelines/minimax_h3_image.md" target="_blank"><svg class="mr-1" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg><!----> <span><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]-->
			
			<script>
				{
					__sveltekit_gxjg5b = {
						base: "/docs/diffusers/main/en",
						assets: "/docs/diffusers/main/en"
					};
					const element = document.currentScript.parentElement;
					Promise.all([
						import("/docs/diffusers/main/en/_app/immutable/entry/start.U6TUchwJ.js"),
						import("/docs/diffusers/main/en/_app/immutable/entry/app.3qKhPoX-.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 9],
							data: [null,null],
							form: null,
							error: null
						});
					});
				}
			</script>
		
Xet Storage Details

Size:: 43.8 kB
Xet hash:: 4ea29c70359907a274df859472dd149ec7875ba0366063ff781359bc633c8e84
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.