Buckets:

hf-doc-build
/

doc

Files
xet
hf-doc-build/doc / diffusers /v0.43.0 /id /using-diffusers /ip_adapter.html
mishig
2 days ago
download
raw
91 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;IP-Adapter&quot;,&quot;local&quot;:&quot;ip-adapter&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Varian checkpoint&quot;,&quot;local&quot;:&quot;checkpoint-variants&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Embedding gambar&quot;,&quot;local&quot;:&quot;image-embeddings&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Masking&quot;,&quot;local&quot;:&quot;masking&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Resep&quot;,&quot;local&quot;:&quot;recipes&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Beberapa IP-Adapter&quot;,&quot;local&quot;:&quot;multiple-ip-adapters&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Kontrol struktural&quot;,&quot;local&quot;:&quot;structural-control&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Kontrol gaya dan tata letak&quot;,&quot;local&quot;:&quot;style-and-layout-control&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Generasi instan&quot;,&quot;local&quot;:&quot;instant-generation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/entry/start.1gaXe03k.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/BObs6oQO.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/DPnyVmXs.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/entry/app.BwdYRbCn.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/DSGUT_OW.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/C045MWWx.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/6tuJ3nyi.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/nodes/0.CornNeQR.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/IaDvMxLQ.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/nodes/321.DEZNj4ih.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/bKdwVYpU.js" rel="modulepreload">
		<link href="/docs/diffusers/v0.43.0/en/_app/immutable/chunks/DTd9IGSp.js" rel="modulepreload">
		<!--1p5pew9--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;IP-Adapter&quot;,&quot;local&quot;:&quot;ip-adapter&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Checkpoint variants&quot;,&quot;local&quot;:&quot;checkpoint-variants&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Image embeddings&quot;,&quot;local&quot;:&quot;image-embeddings&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Masking&quot;,&quot;local&quot;:&quot;masking&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Recipes&quot;,&quot;local&quot;:&quot;recipes&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Multiple IP-Adapters&quot;,&quot;local&quot;:&quot;multiple-ip-adapters&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Structural control&quot;,&quot;local&quot;:&quot;structural-control&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Style and layout control&quot;,&quot;local&quot;:&quot;style-and-layout-control&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3},{&quot;title&quot;:&quot;Instant generation&quot;,&quot;local&quot;:&quot;instant-generation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/><!---->
		<link href="/docs/diffusers/v0.43.0/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="ip-adapter" 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="#ip-adapter"><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><!---->IP-Adapter<!----></span></h1><!--]--><!----> <p><a href="https://huggingface.co/papers/2308.06721" rel="nofollow">IP-Adapter</a> mengarahkan model difusi yang telah dilatih sebelumnya menggunakan gambar referensi sambil tetap mempertahankan prompt teks. Metode ini membekukan model dasar dan menambahkan seperangkat kecil lapisan cross-attention gambar, yang membuatnya praktis untuk mencocokkan subjek atau gaya dari foto tanpa perlu fine-tuning.</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> Disalin</div><!----></button><!----></div> <pre class="language-text "><!---->text prompt                         reference image
    |                                     |
 text encoder                        image encoder
    |                                     |
    v                                     v
 text cross-attn  (frozen)          IP-Adapter cross-attn
    \                                     /
     \                                   /
      +--------&gt;  denoiser  &lt;-----------+
                  (frozen base)<!----></pre></div><!----> <p>Checkpoint IP-Adapter biasanya berukuran ~100MB karena menyimpan bobot adapter, bukan model lengkap. Muat pipeline dasar terlebih dahulu, kemudian muat adapter dengan <a href="/docs/diffusers/v0.43.0/id/api/loaders/ip_adapter#diffusers.loaders.IPAdapterMixin.load_ip_adapter">load_ip_adapter()</a>.</p> <blockquote class="tip"><p>IP-Adapters tersedia untuk banyak model seperti <a href="../api/pipelines/flux#ip-adapter">Flux</a> dan <a href="../api/pipelines/stable_diffusion/stable_diffusion_3">Stable Diffusion 3</a>, serta lainnya. Contoh dalam panduan ini menggunakan Stable Diffusion dan Stable Diffusion XL.</p></blockquote> <p>Gunakan <a href="/docs/diffusers/v0.43.0/id/api/loaders/ip_adapter#diffusers.loaders.IPAdapterMixin.set_ip_adapter_scale">set_ip_adapter_scale()</a> untuk mengontrol seberapa kuat IP-Adapter mengarahkan generasi. <code>1.0</code> menerapkan adapter dengan kekuatan penuh; <code>0.5</code> biasanya menyeimbangkan prompt teks dan gambar.</p> <p>Contoh di bawah ini menunjukkan IP-Adapter pada tugas-tugas umum.</p> <div class="flex space-x-2 items-center my-1.5 mr-8 h-7 !pl-0 -mx-3 md:mx-0"><!--[--><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd border-gray-800 bg-black dark:bg-gray-700 text-white">teks-ke-gambar</div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">image-to-image</div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">inpainting</div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">video</div><!--]--></div> <div class="language-select"><div><div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</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> AutoPipelineForText2Image
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = AutoPipelineForText2Image.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sdxl.bin&quot;</span>
)
pipeline.set_ip_adapter_scale(<span class="hljs-number">0.8</span>)<!----></pre></div><!----> <p>Lewatkan gambar ke <code>ip_adapter_image</code> bersama dengan prompt teks untuk menghasilkan gambar.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_diner.png&quot;</span>)
pipeline(
    prompt=<span class="hljs-string">&quot;a polar bear sitting in a chair drinking a milkshake&quot;</span>,
    ip_adapter_image=image,
    negative_prompt=<span class="hljs-string">&quot;deformed, ugly, wrong proportion, low res, bad anatomy, worst quality, low quality&quot;</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_diner.png" width="400" alt="IP-Adapter image"/> <figcaption style="text-align: center;">Gambar IP-Adapter</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_diner_2.png" width="400" alt="generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></div><!----></div><!----> <div hidden=""><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> Disalin</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> AutoPipelineForImage2Image
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = AutoPipelineForImage2Image.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sdxl.bin&quot;</span>
)
pipeline.set_ip_adapter_scale(<span class="hljs-number">0.8</span>)

image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_bear_1.png&quot;</span>)
ip_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_gummy.png&quot;</span>)
pipeline(
    prompt=<span class="hljs-string">&quot;best quality, high quality&quot;</span>,
    image=image,
    ip_adapter_image=ip_image,
    strength=<span class="hljs-number">0.5</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_bear_1.png" width="300" alt="input image"/> <figcaption style="text-align: center;">gambar input</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_gummy.png" width="300" alt="IP-Adapter image"/> <figcaption style="text-align: center;">Gambar IP-Adapter</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_bear_3.png" width="300" alt="generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></div><!----></div><!----> <div hidden=""><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> Disalin</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> AutoPipelineForInpainting
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = AutoPipelineForInpainting.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sdxl.bin&quot;</span>
)
pipeline.set_ip_adapter_scale(<span class="hljs-number">0.6</span>)

mask_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_mask.png&quot;</span>)
image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_bear_1.png&quot;</span>)
ip_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_gummy.png&quot;</span>)
pipeline(
    prompt=<span class="hljs-string">&quot;a cute gummy bear waving&quot;</span>,
    image=image,
    mask_image=mask_image,
    ip_adapter_image=ip_image,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_bear_1.png" width="300" alt="input image"/> <figcaption style="text-align: center;">gambar input</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_gummy.png" width="300" alt="IP-Adapter image"/> <figcaption style="text-align: center;">Gambar IP-Adapter</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_inpaint.png" width="300" alt="generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></div><!----></div><!----> <div hidden=""><p>Metode <a href="/docs/diffusers/v0.43.0/id/api/pipelines/overview#diffusers.DiffusionPipeline.enable_model_cpu_offload">enable_model_cpu_offload()</a> berguna untuk mengurangi memori dan sebaiknya diaktifkan setelah IP-Adapter dimuat. Jika tidak, encoder gambar IP-Adapter juga akan di-offload ke CPU dan mengembalikan error.</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> Disalin</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> AnimateDiffPipeline, DDIMScheduler, MotionAdapter
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> export_to_gif
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

adapter = MotionAdapter.from_pretrained(
  <span class="hljs-string">&quot;guoyww/animatediff-motion-adapter-v1-5-2&quot;</span>,
  dtype=torch.float16
)
pipeline = AnimateDiffPipeline.from_pretrained(
  <span class="hljs-string">&quot;emilianJR/epiCRealism&quot;</span>,
  motion_adapter=adapter,
  dtype=torch.float16
)
scheduler = DDIMScheduler.from_pretrained(
    <span class="hljs-string">&quot;emilianJR/epiCRealism&quot;</span>,
    subfolder=<span class="hljs-string">&quot;scheduler&quot;</span>,
    clip_sample=<span class="hljs-literal">False</span>,
    timestep_spacing=<span class="hljs-string">&quot;linspace&quot;</span>,
    beta_schedule=<span class="hljs-string">&quot;linear&quot;</span>,
    steps_offset=<span class="hljs-number">1</span>,
)
pipeline.scheduler = scheduler
pipeline.vae.enable_slicing()
pipeline.load_ip_adapter(<span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>, subfolder=<span class="hljs-string">&quot;models&quot;</span>, weight_name=<span class="hljs-string">&quot;ip-adapter_sd15.bin&quot;</span>)
pipeline.enable_model_cpu_offload()

ip_adapter_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_inpaint.png&quot;</span>)
pipeline(
    prompt=<span class="hljs-string">&quot;A cute gummy bear waving&quot;</span>,
    negative_prompt=<span class="hljs-string">&quot;bad quality, worse quality, low resolution&quot;</span>,
    ip_adapter_image=ip_adapter_image,
    num_frames=<span class="hljs-number">16</span>,
    guidance_scale=<span class="hljs-number">7.5</span>,
    num_inference_steps=<span class="hljs-number">50</span>,
).frames[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_inpaint.png" width="400" alt="IP-Adapter image"/> <figcaption style="text-align: center;">Gambar IP-Adapter</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/gummy_bear.gif" width="400" alt="generated video"/> <figcaption style="text-align: center;">video yang dihasilkan</figcaption></figure></div><!----></div><!----><!----></div><!----> <!--[1--><h2 class="relative group"><a id="checkpoint-variants" 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="#checkpoint-variants"><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><!---->Varian checkpoint<!----></span></h2><!--]--><!----> <p>Muat Plus ketika detail dari gambar referensi paling penting. Muat FaceID ketika Anda memerlukan embedding identitas InsightFace alih-alih embedding gambar CLIP.</p> <div class="flex space-x-2 items-center my-1.5 mr-8 h-7 !pl-0 -mx-3 md:mx-0"><!--[--><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd border-gray-800 bg-black dark:bg-gray-700 text-white">IP-Adapter Plus</div><div class="flex items-center border rounded-lg px-1.5 py-1 leading-none select-none text-smd text-gray-500 cursor-pointer opacity-90 hover:text-gray-700 dark:hover:text-gray-200 hover:shadow-sm">IP-Adapter FaceID</div><!--]--></div> <div class="language-select"><div><div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> CLIPVisionModelWithProjection
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> AutoPipelineForText2Image

image_encoder = CLIPVisionModelWithProjection.from_pretrained(
    <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
    subfolder=<span class="hljs-string">&quot;models/image_encoder&quot;</span>,
    dtype=torch.float16
)

pipeline = AutoPipelineForText2Image.from_pretrained(
    <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
    image_encoder=image_encoder,
    dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter-plus_sdxl_vit-h.safetensors&quot;</span>
)<!----></pre></div><!----></div><!----> <div hidden=""><p>Checkpoint FaceID menggunakan face embeddings dari <a href="https://github.com/deepinsight/insightface" rel="nofollow">InsightFace</a> alih-alih image embeddings CLIP. Gunakan <a href="/docs/diffusers/v0.43.0/id/api/schedulers/ddim#diffusers.DDIMScheduler">DDIMScheduler</a> atau <a href="/docs/diffusers/v0.43.0/id/api/schedulers/euler#diffusers.EulerDiscreteScheduler">EulerDiscreteScheduler</a> untuk model FaceID. Ekstrak face embeddings dan lewati sebagai daftar tensor ke <code>ip_adapter_image_embeds</code>.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-comment"># pip install insightface</span>
<span class="hljs-keyword">import</span> cv2
<span class="hljs-keyword">import</span> numpy <span class="hljs-keyword">as</span> np
<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionPipeline, DDIMScheduler
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image
<span class="hljs-keyword">from</span> insightface.app <span class="hljs-keyword">import</span> FaceAnalysis

pipeline = StableDiffusionPipeline.from_pretrained(
    <span class="hljs-string">&quot;stable-diffusion-v1-5/stable-diffusion-v1-5&quot;</span>,
    dtype=torch.float16,
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter-FaceID&quot;</span>,
  subfolder=<span class="hljs-literal">None</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter-faceid_sd15.bin&quot;</span>,
  image_encoder_folder=<span class="hljs-literal">None</span>
)
pipeline.set_ip_adapter_scale(<span class="hljs-number">0.6</span>)

image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_mask_girl1.png&quot;</span>)

ref_images_embeds = []
app = FaceAnalysis(name=<span class="hljs-string">&quot;buffalo_l&quot;</span>, providers=[<span class="hljs-string">&#x27;CUDAExecutionProvider&#x27;</span>, <span class="hljs-string">&#x27;CPUExecutionProvider&#x27;</span>])
app.prepare(ctx_id=<span class="hljs-number">0</span>, det_size=(<span class="hljs-number">640</span>, <span class="hljs-number">640</span>))
image = cv2.cvtColor(np.asarray(image), cv2.COLOR_RGB2BGR)
faces = app.get(image)
image = torch.from_numpy(faces[<span class="hljs-number">0</span>].normed_embedding)
ref_images_embeds.append(image.unsqueeze(<span class="hljs-number">0</span>))
ref_images_embeds = torch.stack(ref_images_embeds, dim=<span class="hljs-number">0</span>).unsqueeze(<span class="hljs-number">0</span>)
neg_ref_images_embeds = torch.zeros_like(ref_images_embeds)
id_embeds = torch.cat([neg_ref_images_embeds, ref_images_embeds]).to(dtype=torch.float16, device=<span class="hljs-string">&quot;cuda&quot;</span>)

pipeline(
    prompt=<span class="hljs-string">&quot;A photo of a girl&quot;</span>,
    ip_adapter_image_embeds=[id_embeds],
    negative_prompt=<span class="hljs-string">&quot;monochrome, lowres, bad anatomy, worst quality, low quality&quot;</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <p>Untuk FaceID Plus, muat <a href="https://huggingface.co/docs/transformers/v5.22.2/en/model_doc/clip#transformers.CLIPVisionModelWithProjection" rel="nofollow">CLIPVisionModelWithProjection</a> sebagai encoder gambar.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> CLIPVisionModelWithProjection
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> AutoPipelineForText2Image

image_encoder = CLIPVisionModelWithProjection.from_pretrained(
    <span class="hljs-string">&quot;laion/CLIP-ViT-H-14-laion2B-s32B-b79K&quot;</span>,
    dtype=torch.float16,
)

pipeline = AutoPipelineForText2Image.from_pretrained(
    <span class="hljs-string">&quot;stable-diffusion-v1-5/stable-diffusion-v1-5&quot;</span>,
    image_encoder=image_encoder,
    dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter-FaceID&quot;</span>,
  subfolder=<span class="hljs-literal">None</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter-faceid-plus_sd15.bin&quot;</span>
)<!----></pre></div><!----> <p>FaceID Plus dan Plus v2 juga memerlukan embedding gambar CLIP di atas embeds InsightFace. Setelah Anda menyiapkan embedding wajah, encode gambar referensi dan tetapkan embeds CLIP ke lapisan proyeksi gambar.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

ip_adapter_images = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_mask_girl1.png&quot;</span>)
num_images = <span class="hljs-number">1</span>
clip_embeds = pipeline.prepare_ip_adapter_image_embeds(
    ip_adapter_image=[ip_adapter_images],
    ip_adapter_image_embeds=<span class="hljs-literal">None</span>,
    device=torch.device(<span class="hljs-string">&quot;cuda&quot;</span>),
    num_images_per_prompt=num_images,
    do_classifier_free_guidance=<span class="hljs-literal">True</span>,
)[<span class="hljs-number">0</span>]

pipeline.unet.encoder_hid_proj.image_projection_layers[<span class="hljs-number">0</span>].clip_embeds = clip_embeds.to(dtype=torch.float16)
<span class="hljs-comment"># set to True if using IP-Adapter FaceID Plus v2</span>
pipeline.unet.encoder_hid_proj.image_projection_layers[<span class="hljs-number">0</span>].shortcut = <span class="hljs-literal">False</span><!----></pre></div><!----><!----></div><!----><!----></div><!----> <!--[1--><h2 class="relative group"><a id="image-embeddings" 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-embeddings"><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><!---->Embedding gambar<!----></span></h2><!--]--><!----> <p><code>prepare_ip_adapter_image_embeds()</code> mengodekan gambar IP-Adapter menjadi embeddings yang dapat Anda simpan dan gunakan kembali. Hitung sebelumnya sekali alih-alih memuat dan mengodekan gambar yang sama setiap kali Anda menjalankan pipeline.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</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> AutoPipelineForText2Image
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = AutoPipelineForText2Image.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sdxl.bin&quot;</span>
)

image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_bear_1.png&quot;</span>)
image_embeds = pipeline.prepare_ip_adapter_image_embeds(
    ip_adapter_image=image,
    ip_adapter_image_embeds=<span class="hljs-literal">None</span>,
    device=<span class="hljs-string">&quot;cuda&quot;</span>,
    num_images_per_prompt=<span class="hljs-number">1</span>,
    do_classifier_free_guidance=<span class="hljs-literal">True</span>,
)

torch.save(image_embeds, <span class="hljs-string">&quot;image_embeds.ipadpt&quot;</span>)<!----></pre></div><!----> <p>Muat ulang embedding gambar dengan meneruskannya ke parameter <code>ip_adapter_image_embeds</code>. Atur <code>image_encoder_folder</code> ke <code>None</code> karena Anda tidak lagi memerlukan image encoder untuk menghasilkan embedding gambar.</p> <blockquote class="tip"><p>Anda juga dapat memuat embedding gambar dari sumber lain seperti ComfyUI.</p></blockquote> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->pipeline = AutoPipelineForText2Image.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  image_encoder_folder=<span class="hljs-literal">None</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sdxl.bin&quot;</span>
)
pipeline.set_ip_adapter_scale(<span class="hljs-number">0.8</span>)
image_embeds = torch.load(<span class="hljs-string">&quot;image_embeds.ipadpt&quot;</span>)
pipeline(
    prompt=<span class="hljs-string">&quot;a polar bear sitting in a chair drinking a milkshake&quot;</span>,
    ip_adapter_image_embeds=image_embeds,
    negative_prompt=<span class="hljs-string">&quot;deformed, ugly, wrong proportion, low res, bad anatomy, worst quality, low quality&quot;</span>,
    num_inference_steps=<span class="hljs-number">100</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="masking" 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="#masking"><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><!---->Masking<!----></span></h2><!--]--><!----> <p>Masking biner memungkinkan penugasan gambar IP-Adapter ke area tertentu pada gambar output, sehingga berguna untuk menyusun beberapa gambar IP-Adapter. Setiap gambar IP-Adapter memerlukan mask biner.</p> <p>Muat <a href="/docs/diffusers/v0.43.0/id/api/loaders/ip_adapter#diffusers.IPAdapterMaskProcessor">IPAdapterMaskProcessor</a> untuk memproses pramasker gambar. Untuk hasil terbaik, berikan <code>height</code> dan <code>width</code> keluaran untuk memastikan mask dengan rasio aspek yang berbeda berukuran sesuai. Jika mask masukan sudah sesuai dengan rasio aspek gambar yang dihasilkan, Anda tidak perlu mengatur <code>height</code> dan <code>width</code>.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</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> AutoPipelineForText2Image
<span class="hljs-keyword">from</span> diffusers.image_processor <span class="hljs-keyword">import</span> IPAdapterMaskProcessor
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = AutoPipelineForText2Image.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

mask1 = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_mask_mask1.png&quot;</span>)
mask2 = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_mask_mask2.png&quot;</span>)

processor = IPAdapterMaskProcessor()
masks = processor.preprocess([mask1, mask2], height=<span class="hljs-number">1024</span>, width=<span class="hljs-number">1024</span>)<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ip_mask_mask1.png" width="200" alt="mask 1"/> <figcaption style="text-align: center;">mask 1</figcaption></figure> <figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ip_mask_mask2.png" width="200" alt="mask 2"/> <figcaption style="text-align: center;">mask 2</figcaption></figure></div> <p>Sediakan gambar IP-Adapter dan skala mereka sebagai daftar. Teruskan mask yang telah dipreproses ke <code>cross_attention_kwargs</code> di pipeline.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->face_image1 = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_mask_girl1.png&quot;</span>)
face_image2 = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_mask_girl2.png&quot;</span>)

pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=[<span class="hljs-string">&quot;ip-adapter-plus-face_sdxl_vit-h.safetensors&quot;</span>]
)
pipeline.set_ip_adapter_scale([[<span class="hljs-number">0.7</span>, <span class="hljs-number">0.7</span>]])

ip_images = [[face_image1, face_image2]]
masks = [masks.reshape(<span class="hljs-number">1</span>, masks.shape[<span class="hljs-number">0</span>], masks.shape[<span class="hljs-number">2</span>], masks.shape[<span class="hljs-number">3</span>])]

pipeline(
  prompt=<span class="hljs-string">&quot;2 girls&quot;</span>,
  ip_adapter_image=ip_images,
  negative_prompt=<span class="hljs-string">&quot;monochrome, lowres, bad anatomy, worst quality, low quality&quot;</span>,
  cross_attention_kwargs={<span class="hljs-string">&quot;ip_adapter_masks&quot;</span>: masks}
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; flex-direction: column; gap: 10px;"><div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ip_mask_girl1.png" width="400" alt="IP-Adapter image 1"/> <figcaption style="text-align: center;">Gambar IP-Adapter 1</figcaption></figure> <figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ip_mask_girl2.png" width="400" alt="IP-Adapter image 2"/> <figcaption style="text-align: center;">IP-Adapter gambar 2</figcaption></figure></div> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_attention_mask_result_seed_0.png" width="400" alt="Generated image with mask"/> <figcaption style="text-align: center;">dihasilkan dengan mask</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_no_attention_mask_result_seed_0.png" width="400" alt="Generated image without mask"/> <figcaption style="text-align: center;">dihasilkan tanpa mask</figcaption></figure></div></div> <!--[1--><h2 class="relative group"><a id="recipes" 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="#recipes"><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><!---->Resep<!----></span></h2><!--]--><!----> <p>Gabungkan IP-Adapter dengan adapter atau pipeline lain, seperti beberapa adapter, ControlNet, InstantStyle, atau LCM few-step, ketika satu gambar referensi tidak cukup.</p> <!--[2--><h3 class="relative group"><a id="multiple-ip-adapters" 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="#multiple-ip-adapters"><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><!---->Beberapa IP-Adapter<!----></span></h3><!--]--><!----> <p>Gabungkan beberapa IP-Adapter untuk menghasilkan gambar dengan gaya yang lebih beragam. Misalnya, Anda dapat menggunakan IP-Adapter Face untuk menghasilkan wajah dan karakter yang konsisten, serta IP-Adapter Plus untuk menghasilkan wajah tersebut dalam gaya tertentu.</p> <p>Muat encoder gambar dengan <a href="https://huggingface.co/docs/transformers/v5.22.2/en/model_doc/clip#transformers.CLIPVisionModelWithProjection" rel="nofollow">CLIPVisionModelWithProjection</a>.</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> Disalin</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> AutoPipelineForText2Image, DDIMScheduler
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> CLIPVisionModelWithProjection
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

image_encoder = CLIPVisionModelWithProjection.from_pretrained(
    <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
    subfolder=<span class="hljs-string">&quot;models/image_encoder&quot;</span>,
    dtype=torch.float16,
)<!----></pre></div><!----> <p>Muat model dasar, penjadwal, dan IP-Adapter berikut.</p> <ul><li><a href="https://huggingface.co/h94/IP-Adapter#ip-adapter-for-sdxl-10" rel="nofollow">ip-adapter-plus_sdxl_vit-h</a> menggunakan patch embeddings dan image encoder ViT-H</li> <li><a href="https://huggingface.co/h94/IP-Adapter#ip-adapter-for-sdxl-10" rel="nofollow">ip-adapter-plus-face_sdxl_vit-h</a> menggunakan patch embeddings dan image encoder ViT-H, tetapi dikondisikan pada gambar wajah yang dipotong</li></ul> <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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->pipeline = AutoPipelineForText2Image.from_pretrained(
    <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
    dtype=torch.float16,
    image_encoder=image_encoder,
)
pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=[<span class="hljs-string">&quot;ip-adapter-plus_sdxl_vit-h.safetensors&quot;</span>, <span class="hljs-string">&quot;ip-adapter-plus-face_sdxl_vit-h.safetensors&quot;</span>]
)
pipeline.set_ip_adapter_scale([<span class="hljs-number">0.7</span>, <span class="hljs-number">0.3</span>])
<span class="hljs-comment"># enable_model_cpu_offload to reduce memory usage</span>
pipeline.enable_model_cpu_offload()<!----></pre></div><!----> <p>Muat sebuah gambar dan folder yang berisi gambar-gambar dengan gaya tertentu untuk diterapkan.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->face_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/women_input.png&quot;</span>)
style_folder = <span class="hljs-string">&quot;https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/style_ziggy&quot;</span>
style_images = [load_image(<span class="hljs-string">f&quot;<span class="hljs-subst">{style_folder}</span>/img<span class="hljs-subst">{i}</span>.png&quot;</span>) <span class="hljs-keyword">for</span> i <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(<span class="hljs-number">10</span>)]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/women_input.png" width="400" alt="Face image"/> <figcaption style="text-align: center;">gambar wajah</figcaption></figure> <figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ip_style_grid.png" width="400" alt="Style images"/> <figcaption style="text-align: center;">gaya gambar</figcaption></figure></div> <p>Lewatkan gambar gaya dan wajah sebagai daftar ke <code>ip_adapter_image</code>.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->generator = torch.Generator(device=<span class="hljs-string">&quot;cpu&quot;</span>).manual_seed(<span class="hljs-number">0</span>)

pipeline(
    prompt=<span class="hljs-string">&quot;wonderwoman&quot;</span>,
    ip_adapter_image=[style_images, face_image],
    negative_prompt=<span class="hljs-string">&quot;monochrome, lowres, bad anatomy, worst quality, low quality&quot;</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; justify-content: center;"><figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ip_multi_out.png" width="400" alt="Generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></div> <!--[2--><h3 class="relative group"><a id="structural-control" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#structural-control"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Kontrol struktural<!----></span></h3><!--]--><!----> <p>Untuk kontrol struktural, kombinasikan IP-Adapter dengan <a href="../api/pipelines/controlnet">ControlNet</a> yang dikondisikan pada peta kedalaman, peta tepi, estimasi pose, dan lainnya.</p> <p>Contoh di bawah ini memuat checkpoint <a href="/docs/diffusers/v0.43.0/id/api/models/controlnet#diffusers.ControlNetModel">ControlNetModel</a> yang dikondisikan pada peta kedalaman dan menggabungkannya dengan IP-Adapter.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionControlNetPipeline, ControlNetModel

controlnet = ControlNetModel.from_pretrained(
  <span class="hljs-string">&quot;lllyasviel/control_v11f1p_sd15_depth&quot;</span>,
  dtype=torch.float16
)

pipeline = StableDiffusionControlNetPipeline.from_pretrained(
    <span class="hljs-string">&quot;stable-diffusion-v1-5/stable-diffusion-v1-5&quot;</span>,
    controlnet=controlnet,
    dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sd15.bin&quot;</span>
)<!----></pre></div><!----> <p>Lewatkan peta kedalaman dan gambar IP-Adapter ke pipeline.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->depth_map = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/depth.png&quot;</span>)
ip_adapter_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/statue.png&quot;</span>)
pipeline(
  prompt=<span class="hljs-string">&quot;best quality, high quality&quot;</span>,
  image=depth_map,
  ip_adapter_image=ip_adapter_image,
  negative_prompt=<span class="hljs-string">&quot;monochrome, lowres, bad anatomy, worst quality, low quality&quot;</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/statue.png" width="300" alt="IP-Adapter image"/> <figcaption style="text-align: center;">Gambar IP-Adapter</figcaption></figure> <figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/depth.png" width="300" alt="Depth map"/> <figcaption style="text-align: center;">peta kedalaman</figcaption></figure> <figure><img src="https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/ipa-controlnet-out.png" width="300" alt="Generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></div> <!--[2--><h3 class="relative group"><a id="style-and-layout-control" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#style-and-layout-control"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Kontrol gaya dan tata letak<!----></span></h3><!--]--><!----> <p>Untuk kontrol gaya dan tata letak, kombinasikan IP-Adapter dengan <a href="https://huggingface.co/papers/2404.02733" rel="nofollow">InstantStyle</a>. InstantStyle memisahkan <em>gaya</em> (warna, tekstur, kesan keseluruhan) dari <em>konten</em> dan menerapkan gaya hanya pada blok yang spesifik untuk gaya, sehingga area konten tetap utuh. Hal ini memberikan konsistensi gaya yang lebih kuat dan kontrol tata letak yang lebih jelas.</p> <p>Aktifkan IP-Adapter hanya pada lapisan tertentu dengan <a href="/docs/diffusers/v0.43.0/id/api/loaders/ip_adapter#diffusers.loaders.IPAdapterMixin.set_ip_adapter_scale">set_ip_adapter_scale()</a>. Contoh di bawah ini mengaktifkannya pada <code>block_2</code> (layout) bagian down dan <code>block_0</code> (style) bagian up dari model.</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> Disalin</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> AutoPipelineForText2Image
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = AutoPipelineForText2Image.from_pretrained(
  <span class="hljs-string">&quot;stabilityai/stable-diffusion-xl-base-1.0&quot;</span>,
  dtype=torch.float16
).to(<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;sdxl_models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sdxl.bin&quot;</span>
)

scale = {
    <span class="hljs-string">&quot;down&quot;</span>: {<span class="hljs-string">&quot;block_2&quot;</span>: [<span class="hljs-number">0.0</span>, <span class="hljs-number">1.0</span>]},
    <span class="hljs-string">&quot;up&quot;</span>: {<span class="hljs-string">&quot;block_0&quot;</span>: [<span class="hljs-number">0.0</span>, <span class="hljs-number">1.0</span>, <span class="hljs-number">0.0</span>]},
}
pipeline.set_ip_adapter_scale(scale)<!----></pre></div><!----> <p>Muat gambar gaya dan hasilkan gambar.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->style_image = load_image(<span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg&quot;</span>)

pipeline(
    prompt=<span class="hljs-string">&quot;a cat, masterpiece, best quality, high quality&quot;</span>,
    ip_adapter_image=style_image,
    negative_prompt=<span class="hljs-string">&quot;text, watermark, lowres, low quality, worst quality, deformed, glitch, low contrast, noisy, saturation, blurry&quot;</span>,
    guidance_scale=<span class="hljs-number">5</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg" width="400" alt="Style image"/> <figcaption style="text-align: center;">style image</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/datasets/cat_style_layout.png" width="400" alt="Generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></div> <p>Gambar-gambar di bawah ini membandingkan aktivasi gaya-only InstantStyle (hanya <code>block_0</code>) dengan mengaktifkan IP-Adapter di semua lapisan. Pengaturan semua lapisan biasanya mengikuti prompt gambar lebih kuat dan dapat mengurangi keragaman. Pilih skala gaya-only jika Anda menginginkan perilaku pelestarian tata letak dari InstantStyle.</p> <blockquote class="tip"><p>Anda tidak perlu menentukan semua lapisan dalam kamus <code>scale</code>. Lapisan yang tidak disertakan diatur ke 0, yang berarti IP-Adapter dinonaktifkan.</p></blockquote> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600  " title="code excerpt" type="button"><svg xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg><!----> <div class=" absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0 "><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent;"></div> Disalin</div><!----></button><!----></div> <pre class="language-py "><!---->scale = {
    <span class="hljs-string">&quot;up&quot;</span>: {<span class="hljs-string">&quot;block_0&quot;</span>: [<span class="hljs-number">0.0</span>, <span class="hljs-number">1.0</span>, <span class="hljs-number">0.0</span>]},
}
pipeline.set_ip_adapter_scale(scale)

pipeline(
    prompt=<span class="hljs-string">&quot;a cat, masterpiece, best quality, high quality&quot;</span>,
    ip_adapter_image=style_image,
    negative_prompt=<span class="hljs-string">&quot;text, watermark, lowres, low quality, worst quality, deformed, glitch, low contrast, noisy, saturation, blurry&quot;</span>,
    guidance_scale=<span class="hljs-number">5</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; gap: 10px; justify-content: space-around; align-items: flex-end;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/datasets/cat_style_only.png" width="400" alt="Generated image (style only)"/> <figcaption style="text-align: center;">hanya-gaya (blok_0 atas)</figcaption></figure> <figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/datasets/cat_ip_adapter.png" width="400" alt="Generated image (all layers)"/> <figcaption style="text-align: center;">semua lapisan</figcaption></figure></div> <!--[2--><h3 class="relative group"><a id="instant-generation" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#instant-generation"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Generasi instan<!----></span></h3><!--]--><!----> <p>Gabungkan IP-Adapter dengan <a href="../api/pipelines/latent_consistency_models">LCM</a> LoRA untuk generasi beberapa langkah.</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> Disalin</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline, LCMScheduler
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image

pipeline = DiffusionPipeline.from_pretrained(
  <span class="hljs-string">&quot;sd-dreambooth-library/herge-style&quot;</span>,
  dtype=torch.float16
)
pipeline.load_ip_adapter(
  <span class="hljs-string">&quot;h94/IP-Adapter&quot;</span>,
  subfolder=<span class="hljs-string">&quot;models&quot;</span>,
  weight_name=<span class="hljs-string">&quot;ip-adapter_sd15.bin&quot;</span>
)
pipeline.load_lora_weights(<span class="hljs-string">&quot;latent-consistency/lcm-lora-sdv1-5&quot;</span>)
pipeline.scheduler = LCMScheduler.from_config(pipeline.scheduler.config)
pipeline.enable_model_cpu_offload()

pipeline.set_ip_adapter_scale(<span class="hljs-number">0.4</span>)
ip_adapter_image = load_image(<span class="hljs-string">&quot;https://user-images.githubusercontent.com/24734142/266492875-2d50d223-8475-44f0-a7c6-08b51cb53572.png&quot;</span>)
pipeline(
    prompt=<span class="hljs-string">&quot;herge_style woman in armor, best quality, high quality&quot;</span>,
    ip_adapter_image=ip_adapter_image,
    num_inference_steps=<span class="hljs-number">4</span>,
    guidance_scale=<span class="hljs-number">1</span>,
).images[<span class="hljs-number">0</span>]<!----></pre></div><!----> <div style="display: flex; justify-content: center;"><figure><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/ip_adapter_herge.png" width="400" alt="Generated image"/> <figcaption style="text-align: center;">gambar yang dihasilkan</figcaption></figure></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/using-diffusers/ip_adapter.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_ywq9g6 = {
						base: "/docs/diffusers/v0.43.0/en",
						assets: "/docs/diffusers/v0.43.0/en"
					};
					const element = document.currentScript.parentElement;
					Promise.all([
						import("/docs/diffusers/v0.43.0/en/_app/immutable/entry/start.1gaXe03k.js"),
						import("/docs/diffusers/v0.43.0/en/_app/immutable/entry/app.BwdYRbCn.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 321],
							data: [null,null],
							form: null,
							error: null
						});
					});
				}
			</script>
		
Xet Storage Details

Size:: 91 kB
Xet hash:: 389484f38bbdaa6793103368ad64afd43ec3241b2a0605cdc91bb89e67f4326e
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.