Buckets:

hf-doc-build
/

doc

Files
xet
hf-doc-build/doc / diffusers /main /tr /api /pipelines /pika_h3_flex.html
mishig
about 15 hours ago
download
raw
109 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Pika H3 Flex&quot;,&quot;local&quot;:&quot;pika-h3-flex&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Yükleniyor&quot;,&quot;local&quot;:&quot;loading&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Yüklenen MiniMax-H3'ün yeniden kullanımı&quot;,&quot;local&quot;:&quot;reusing-a-loaded-minimax-h3&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2},{&quot;title&quot;:&quot;MiniMax-H3'ten farklı olan&quot;,&quot;local&quot;:&quot;what-differs-from-minimax-h3&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Üretim&quot;,&quot;local&quot;:&quot;generation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.PikaH3FlexModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexBlocks&quot;,&quot;local&quot;:&quot;diffusers.PikaH3FlexBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexImageReference&quot;,&quot;local&quot;:&quot;diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexVideoReference&quot;,&quot;local&quot;:&quot;diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexAudioReference&quot;,&quot;local&quot;:&quot;diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
		<link href="/docs/diffusers/main/en/_app/immutable/entry/start.vZc-dOUR.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Bafuik5c.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/Bupkjr9f.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/entry/app.B6c3Ml05.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/CyvZ3FS5.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/l0QnnOED.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/nodes/0.BQlx092V.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/CoSrdQax.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/nodes/223.fCsW7WyL.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/B4AMoTAu.js" rel="modulepreload">
		<link href="/docs/diffusers/main/en/_app/immutable/chunks/BCPzXmeI.js" rel="modulepreload">
		<!--13wh65h--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Pika H3 Flex&quot;,&quot;local&quot;:&quot;pika-h3-flex&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Loading&quot;,&quot;local&quot;:&quot;loading&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Reusing a loaded MiniMax-H3&quot;,&quot;local&quot;:&quot;reusing-a-loaded-minimax-h3&quot;,&quot;sections&quot;:[],&quot;depth&quot;:3}],&quot;depth&quot;:2},{&quot;title&quot;:&quot;What differs from MiniMax-H3&quot;,&quot;local&quot;:&quot;what-differs-from-minimax-h3&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Generation&quot;,&quot;local&quot;:&quot;generation&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexModularPipeline&quot;,&quot;local&quot;:&quot;diffusers.PikaH3FlexModularPipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexBlocks&quot;,&quot;local&quot;:&quot;diffusers.PikaH3FlexBlocks&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexImageReference&quot;,&quot;local&quot;:&quot;diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexVideoReference&quot;,&quot;local&quot;:&quot;diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;PikaH3FlexAudioReference&quot;,&quot;local&quot;:&quot;diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference&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="pika-h3-flex" 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="#pika-h3-flex"><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><!---->Pika H3 Flex<!----></span></h1><!--]--><!----> <div class="flex flex-wrap space-x-1"><a href="https://huggingface.co/docs/diffusers/main/en/tutorials/using_peft_for_inference" target="_blank" rel="noopener"><img alt="LoRA" src="https://img.shields.io/badge/LoRA-d8b4fe?style=flat"/></a></div> <p><a href="https://huggingface.co/Pika-Labs/Pika-H3-Flex" rel="nofollow">Pika H3 Flex</a>, <a href="./minimax_h3">MiniMax-H3</a>'ün altı adımlı bir video ve ses üreticisine dönüştürülmesi için Pika tarafından yapılan eğitim sonrası (post-training) işlemidir. MiniMax-H3'ün sıkıştırılmış dizisini (packed sequence), Qwen3-VL koşullandırıcısını (conditioner) ve iki VAE'sini korur; iki şeyi değiştirir: örnekleyici (sampler), Euler güncellemesi yerine her adımda temiz tahmini yerinde yeniden gürültüleyen (re-noise) sabit bir altı adımlık zaman çizelgesi ve ağırlıklar, iki MiniMax-H3 dönüştürücü (transformer) bölümü üzerinde bir çift rank-128 LoRA adaptörü. Video ve onun senkronize stereo ses bandı, aynı altı ileri geçişten (forward pass) çıkar.</p> <p>Sürüm, adaptörlerdir: FL2VA adaptörü (<code>fl2va/adapter.safetensors</code>), metin ve uç kare koşullandırması için <code>transformer/</code> bölümünü sürerken, Ref2VA adaptörü (<code>ref2va/adapter.safetensors</code>), çok modlu referanslar için <code>transformer_ref/</code> bölümünü sürer. Temel bileşenler <a href="https://huggingface.co/MiniMaxAI/MiniMax-H3" rel="nofollow">MiniMaxAI/MiniMax-H3</a>'e aittir ve değiştirilmemiştir; ayrıca, repozituar kendi başına yüklenebilmesi için bu bileşenlerin bir kopyasını barındırır.</p> <p>Pika H3 Flex, MiniMax-H3 gibi yalnızca <a href="../../modular_diffusers/overview">Modular Diffusers</a> blokları olarak entegre edilir: bloklar ve <a href="/docs/diffusers/main/tr/api/pipelines/pika_h3_flex#diffusers.PikaH3FlexModularPipeline">PikaH3FlexModularPipeline</a> tüm entegrasyonu oluşturur ve <code>DiffusionPipeline</code> tarafı yoktur.</p> <!--[1--><h2 class="relative group"><a id="loading" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#loading"><span><svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg><!----></span></a> <span><!---->Yükleniyor<!----></span></h2><!--]--><!----> <p>Bir pipeline iki adımda oluşturulur: bileşenler, MiniMax-H3 temel bileşenlerini ve iki zamanlayıcı yapılandırmasını adlandıran <code>modular_model_index.json</code> üzerinden yüklenir, ardından seçilen iş akışının adaptörü <a href="/docs/diffusers/main/tr/api/loaders/lora#diffusers.loaders.MiniMaxH3LoraLoaderMixin.load_lora_weights">load_lora_weights()</a> ile üzerine yüklenir. Adaptörler, diffusers modül adlarının düz PEFT dökümleridir ve eğitim ölçeklerinde yüklenir (<code>alpha == rank</code>, 1.0).</p> <table><thead><tr><th>İş Akışı</th><th>Girdiler</th><th>Dönüştürücü bölme</th><th>Uyarlama</th></tr></thead><tbody><tr><td><code>t2va</code></td><td><code>prompt</code></td><td><code>transformer/</code></td><td><code>fl2va/adapter.safetensors</code></td></tr><tr><td><code>fl2va</code></td><td><code>prompt</code> ve <code>image</code> ve/veya <code>last_image</code></td><td><code>transformer/</code></td><td><code>fl2va/adapter.safetensors</code></td></tr><tr><td><code>ref2va</code></td><td><code>prompt</code> ve <code>references</code></td><td><code>transformer_ref/</code></td><td><code>ref2va/adapter.safetensors</code></td></tr></tbody></table> <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> Kopyalandı</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;Pika-Labs/Pika-H3-Flex&quot;</span>, workflow=<span class="hljs-string">&quot;fl2va&quot;</span>)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;audio_vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
pipe.load_lora_weights(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, weight_name=<span class="hljs-string">&quot;fl2va/adapter.safetensors&quot;</span>)<!----></pre></div><!----> <p>Workflow parametresinin geçirilmesi, blokları o göreve göre budar; böylece yalnızca kendi transformer bölümü alınır, adaptör mevcut tek bölüme yönlendirilir ve pipeline yalnızca o görevin isteklerini kabul eder — <code>fl2va</code> pipeline'ı bir anahtar kare gerektirirken, <code>t2va</code> pipeline'ı hiçbir şey almaz. Her iki bölümü de barındıran (bir <code>workflow=</code> olmadan yüklenen) bir pipeline için, Ref2VA adaptörünün nereye gideceği belirtilmelidir: <code>pipe.load_lora_weights(..., weight_name="ref2va/adapter.safetensors", load_into_transformer_ref=True)</code>. İki VAE, yayınlanan paketin yaptığı gibi float32 olarak yayınlanır ve float32'de çözümlenir; her şey bfloat16'da çalışır.</p> <blockquote class="tip"><p><code>pipe.doc</code>, önünüzdeki hattın ne aldığını ve ne döndürdüğünü gösterir: varsayılanlarıyla birlikte her girdi, beklediği bileşenler ve ürettiği çıktılar.</p></blockquote> <!--[2--><h3 class="relative group"><a id="reusing-a-loaded-minimax-h3" 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="#reusing-a-loaded-minimax-h3"><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><!---->Yüklenen MiniMax-H3'ün yeniden kullanımı<!----></span></h3><!--]--><!----> <p>Temel bileşenler MiniMax-H3’ün olduğundan, zaten yüklenmiş bir MiniMax-H3 hattı, iki zamanlayıcı ve adaptörler dışında her şeyi barındırır. Bu bölüm, yüklenmiş bu bileşenlerden bir Pika H3 Flex hattı oluşturur ve her iki hattı yan yana hazır tutar: tek bir <a href="/docs/diffusers/main/tr/api/modular_diffusers/pipeline_components#diffusers.ComponentsManager">ComponentsManager</a>, her bileşeni bir kez tutar ve her iki hat için de dışa aktarır; Pika H3 Flex hattı yalnızca kendi iki zamanlayıcısını ekler ve adaptörler, paylaşılan dönüştürücülerin iki modelden hangisini çalıştıracağını belirler.</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> Kopyalandı</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

dtype = {<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;audio_vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16}
manager = ComponentsManager()
manager.enable_auto_cpu_offload(device=<span class="hljs-string">&quot;cuda&quot;</span>)

<span class="hljs-comment"># The MiniMax-H3 pipeline already in use, with every workflow loaded. `collection` registers its components in the</span>
<span class="hljs-comment"># manager under that name, so they can be fetched by it later.</span>
minimax = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;MiniMaxAI/MiniMax-H3&quot;</span>, components_manager=manager, collection=<span class="hljs-string">&quot;minimax-h3&quot;</span>)
minimax.load_components(dtype=dtype)

<span class="hljs-comment"># The Pika H3 Flex pipeline, in its own collection. Only its own components are loaded: the two schedulers.</span>
pika = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, components_manager=manager, collection=<span class="hljs-string">&quot;pika-h3-flex&quot;</span>)
pika.load_components(names=[<span class="hljs-string">&quot;scheduler&quot;</span>, <span class="hljs-string">&quot;audio_scheduler&quot;</span>])

<span class="hljs-comment"># The rest comes from the MiniMax-H3 collection, and `update_components` puts it in the Pika H3 Flex pipeline.</span>
shared = [<span class="hljs-string">&quot;text_encoder&quot;</span>, <span class="hljs-string">&quot;tokenizer&quot;</span>, <span class="hljs-string">&quot;processor&quot;</span>, <span class="hljs-string">&quot;vae&quot;</span>, <span class="hljs-string">&quot;audio_vae&quot;</span>, <span class="hljs-string">&quot;transformer&quot;</span>, <span class="hljs-string">&quot;transformer_ref&quot;</span>]
pika.update_components(**{name: manager.get_one(name=name, collection=<span class="hljs-string">&quot;minimax-h3&quot;</span>) <span class="hljs-keyword">for</span> name <span class="hljs-keyword">in</span> shared})

<span class="hljs-comment"># The adapters go into the shared transformers.</span>
pika.load_lora_weights(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, weight_name=<span class="hljs-string">&quot;fl2va/adapter.safetensors&quot;</span>)
pika.load_lora_weights(
    <span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, weight_name=<span class="hljs-string">&quot;ref2va/adapter.safetensors&quot;</span>, load_into_transformer_ref=<span class="hljs-literal">True</span>
)<!----></pre></div><!----> <p><code>print(manager)</code> sonucu şu şekilde gösterilir: her paylaşılan bileşen bir kez listelenir ve her iki koleksiyona da aittir, Pika H3 Flex koleksiyonu iki zamanlayıcısını ekler ve iki dönüştürücü bir adaptör taşır.</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> Kopyalandı</div><!----></button><!----></div> <pre class="language-text "><!---->Models:
Name_ID               | Class                           | Size (GB) | Collection
transformer_1397...   | MiniMaxH3Transformer3DModel     | 63.02     | minimax-h3
                      |                                 |           | pika-h3-flex
transformer_ref_13... | MiniMaxH3Transformer3DModel     | 63.02     | minimax-h3
                      |                                 |           | pika-h3-flex
text_encoder_1397...  | Qwen3VLForConditionalGeneration | 62.13     | minimax-h3
                      |                                 |           | pika-h3-flex
vae_1397...           | AutoencoderKLMiniMaxH3          | 9.70      | minimax-h3
                      |                                 |           | pika-h3-flex
audio_vae_1397...     | AutoencoderKLMiniMaxH3Audio     | 0.56      | minimax-h3
                      |                                 |           | pika-h3-flex

Other Components:
ID                        | Class              | Collection
scheduler_1397...         | MiniMaxH3Scheduler | minimax-h3
audio_scheduler_1397...   | MiniMaxH3Scheduler | minimax-h3
scheduler_1397...         | MiniMaxH3Scheduler | pika-h3-flex
audio_scheduler_1397...   | MiniMaxH3Scheduler | pika-h3-flex
tokenizer_1397...         | Qwen2Tokenizer     | minimax-h3
                          |                    | pika-h3-flex
processor_1397...         | Qwen3VLProcessor   | minimax-h3
                          |                    | pika-h3-flex

Additional Component Info:
transformer_1397...:
  Adapters: [&#x27;default_0&#x27;]
transformer_ref_13...:
  Adapters: [&#x27;default_0&#x27;]<!----></pre></div><!----> <p>Adaptörler, paylaşılan dönüştürücülerin hangi model olduğunu belirler. MiniMax-H3'e geri dönmek için <code>pika.disable_lora()</code> çağrısı yapmanız yeterlidir: sonraki <code>minimax(...)</code> isteği temel modeli çalıştırır. <code>pika.enable_lora()</code> Pika H3 Flex'i yeniden etkinleştirir. Arada hiçbir şey yeniden yüklenmez; yönetici, her iki deyiş hattı da ilgili bileşene ulaştığında her bir bileşeni hızlandırıcı üzerinde açıp kapatır.</p> <!--[1--><h2 class="relative group"><a id="what-differs-from-minimax-h3" 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="#what-differs-from-minimax-h3"><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'ten farklı olan<!----></span></h2><!--]--><!----> <ul><li><strong>Altı adım, <code>num_inference_steps</code> yok.</strong> Her iki <a href="/docs/diffusers/main/tr/api/schedulers/minimax_h3#diffusers.MiniMaxH3Scheduler">MiniMaxH3Scheduler</a> örneği de (video için <code>shift=12.0</code>, ses için <code>shift=3.0</code>), hattın <code>denoising_step_list</code> yapılandırması olan yayınlanan <code>[1000, 750, 500, 250, 100, 40, 0]</code> zamanlaması üzerinde <code>stochastic_sampling</code> ile çalışır: her adım temiz örneği tahmin eder ve sondan bir önceki hariç tüm adımlar, onu bir sonraki sigmaya yeni gürültüyle yeniden gürültüler. Adım sayısı, adaptörün sözleşmesidir ve bir girdi değildir.</li> <li><strong>Her adımda taze gürültü.</strong> Bir istek, <code>generator</code> içinden sabit bir sırayla çekim yapar: koşullandırma gürültüsü için her anahtar kare veya görsel referans için bir çekim, video gürültüsü, ses gürültüsü ve ardından ilk beş adımın her birindeki video ve ses yeniden gürültüleme çekimleri. <code>latents</code> veya <code>audio_latents</code> geçirildiğinde, ilk iki çekim değiştirilir ve geri kalanı olduğu gibi kalır.</li> <li><strong>Keyframe'lar kırpılır, uzatılmaz.</strong> <code>fl2va</code> keyframe'i, kendi çözünürlüğünde tuvalin en-boy oranına göre merkezden kırpılır ve tuvale yeniden boyutlandırılır.</li> <li><strong>Referanslar tuvalin çözünürlük katmanını takip eder.</strong> Pika H3 Flex, bir sepet tablosu üzerinde üretim yapar — 480p ve 768p katmanları, her biri on bir en-boy oranı içerir — ve bir <code>ref2va</code> isteği, tuvalinin katmanına göre koşullandırılır: bir görüntü referansı, 768 katmanının altında 1024 piksel kısa kenar ve bu katmandan yukarıda 2048 piksel olarak kodlanır; bir video referansı ise katmanın kendi tuval kuralına göre yerleştirilir. Her video veya bağımsız ses referansı 2 ila 15 saniye arasında sürer ve video süreleri ile bağımsız ses süreleri ayrı ayrı en fazla 15 saniyeyi bulur.</li></ul> <!--[1--><h2 class="relative group"><a id="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="#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><!---->Üretim<!----></span></h2><!--]--><!----> <p>Yayınlanan kovalar, Pika H3 Flex'in doğrulandığı tuval boyutlarıdır. Bunlardan birini <code>height</code> ve <code>width</code> olarak geçirin:</p> <table><thead><tr><th></th><th>21:9</th><th>16:9</th><th>3:2</th><th>4:3</th><th>5:4</th><th>1:1</th><th>4:5</th><th>3:4</th><th>2:3</th><th>9:16</th><th>9:21</th></tr></thead><tbody><tr><td>480p</td><td>960x416</td><td>832x480</td><td>704x480</td><td>640x480</td><td>608x480</td><td>480x480</td><td>480x608</td><td>480x640</td><td>480x704</td><td>480x832</td><td>416x960</td></tr><tr><td>768p</td><td>1536x672</td><td>1344x768</td><td>1152x768</td><td>1024x768</td><td>960x768</td><td>768x768</td><td>768x960</td><td>768x1024</td><td>768x1152</td><td>768x1344</td><td>672x1536</td></tr></tbody></table> <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> Kopyalandı</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
<span class="hljs-keyword">from</span> diffusers.utils <span class="hljs-keyword">import</span> load_image
<span class="hljs-keyword">from</span> diffusers.utils.export_utils <span class="hljs-keyword">import</span> encode_video

<span class="hljs-comment"># 61.7GB of transformer and 62.1GB of conditioner do not sit on one accelerator, so the components are</span>
<span class="hljs-comment"># registered in a manager that moves each one on and off as the blocks reach it.</span>
manager = ComponentsManager()
manager.enable_auto_cpu_offload(device=<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

<span class="hljs-comment"># Create one pipeline for all supported workflows, then load the `fl2va` workflow&#x27;s components, which is</span>
<span class="hljs-comment"># enough to run both the fl2va and t2va requests below.</span>
pipe = ModularPipeline.from_pretrained(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, components_manager=manager)
pipe.load_components(
    workflow=<span class="hljs-string">&quot;fl2va&quot;</span>, dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;audio_vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16}
)
pipe.load_lora_weights(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, weight_name=<span class="hljs-string">&quot;fl2va/adapter.safetensors&quot;</span>)

prompt = (
    <span class="hljs-string">&quot;integrated_multimodal_description: A cinematic close shot of a red ceramic cup on a wooden table beside an &quot;</span>
    <span class="hljs-string">&quot;open window. A soft breeze moves a linen curtain. The camera slowly pushes toward the cup as warm morning &quot;</span>
    <span class="hljs-string">&quot;light crosses the table.\n&quot;</span>
    <span class="hljs-string">&quot;overall_soundscape: Gentle wind through the window and faint birdsong outside. No speech.\n&quot;</span>
    <span class="hljs-string">&quot;non_diegetic_music: None.\n&quot;</span>
)
outputs = [<span class="hljs-string">&quot;videos&quot;</span>, <span class="hljs-string">&quot;audio&quot;</span>, <span class="hljs-string">&quot;sampling_rate&quot;</span>]

<span class="hljs-comment"># Text to video + audio on the 480p 16:9 bucket, 5 seconds.</span>
results = pipe(
    prompt=prompt, 
    height=<span class="hljs-number">480</span>, 
    width=<span class="hljs-number">832</span>, 
    num_frames=<span class="hljs-number">124</span>, 
    generator=torch.Generator(<span class="hljs-string">&quot;cuda&quot;</span>).manual_seed(<span class="hljs-number">42</span>), 
    output=outputs
)

<span class="hljs-comment"># First frame (and optionally `last_image`) to video + audio. The keyframe is center-cropped onto the canvas and</span>
<span class="hljs-comment"># held as the first frame, so the prompt describes the keyframe and what happens from it.</span>
image = load_image(
    <span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/astronaut.jpg&quot;</span>
)
i2va_prompt = (
    <span class="hljs-string">&quot;For the target video, at 0.00 seconds into the target video, &lt;Picture 1&gt; (from [Shot 1]) is fully referenced.\n\n&quot;</span>
    <span class="hljs-string">&quot;integrated_multimodal_description: [Shot 1] Begin with the supplied image of a small toy astronaut figurine &quot;</span>
    <span class="hljs-string">&quot;standing on a grey rocky moon surface, holding a white egg-shaped object. Preserve the figurine, its orange &quot;</span>
    <span class="hljs-string">&quot;visor and the rocks. The astronaut slowly turns its helmet toward the camera and lifts the egg a little; fine &quot;</span>
    <span class="hljs-string">&quot;dust drifts up around its boots. The camera slowly pushes in in one continuous shot. By the end the astronaut &quot;</span>
    <span class="hljs-string">&quot;stands still, egg held up.\n\n&quot;</span>
    <span class="hljs-string">&quot;overall_soundscape: Soft low wind and faint crunch of dust under the boots. No speech.\n\n&quot;</span>
    <span class="hljs-string">&quot;non_diegetic_music: N/A\n&quot;</span>
)
results = pipe(
    prompt=i2va_prompt, 
    image=image, 
    height=<span class="hljs-number">480</span>, 
    width=<span class="hljs-number">832</span>, 
    num_frames=<span class="hljs-number">124</span>,
    generator=torch.Generator(<span class="hljs-string">&quot;cuda&quot;</span>).manual_seed(<span class="hljs-number">42</span>), 
    output=outputs,
)

encode_video(
    results[<span class="hljs-string">&quot;videos&quot;</span>][<span class="hljs-number">0</span>],
    fps=<span class="hljs-number">24</span>,
    output_path=<span class="hljs-string">&quot;pika_h3_flex_fl2va.mp4&quot;</span>,
    audio=results[<span class="hljs-string">&quot;audio&quot;</span>][<span class="hljs-number">0</span>],
    audio_sample_rate=results[<span class="hljs-string">&quot;sampling_rate&quot;</span>],
)<!----></pre></div><!----> <p><code>ref2va</code> iş akışı, MiniMax-H3’ün yaptığı gibi, sıralı bir referans listesi alır — en fazla 9 görüntü, 3 video ve 3 ses kliği, toplamda 12 — her modallite için bir referans sınıfı ile. Ayrıca en çok bellek gerektiren istektir: referanslar dizinin kendi satırlarını ekler ve 768p’de yalnızca transformer 80 GB’lık bir kartı doldurur. Bu nedenle aşağıdaki örnek, transformer’ın bloklarını CPU belleğinden birer birer akış olarak işlerken, yönetici diğer bileşenleri önceki gibi cihaz üzerinde ve dışında taşır. Tek bir 80 GB’lık kart üzerinde 768p referans isteği çalıştırır; akış olmadan 480p’de kalın.</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> Kopyalandı</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
<span class="hljs-keyword">from</span> diffusers.modular_pipelines.pika_h3_flex <span class="hljs-keyword">import</span> (
    PikaH3FlexAudioReference,
    PikaH3FlexImageReference,
    PikaH3FlexVideoReference,
)

manager = ComponentsManager()
manager.enable_auto_cpu_offload(device=<span class="hljs-string">&quot;cuda&quot;</span>)  <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>

pipe = ModularPipeline.from_pretrained(
    <span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, workflow=<span class="hljs-string">&quot;ref2va&quot;</span>, components_manager=manager
)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;audio_vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
pipe.load_lora_weights(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, weight_name=<span class="hljs-string">&quot;ref2va/adapter.safetensors&quot;</span>)

<span class="hljs-comment"># Keep the transformer in CPU memory and bring in one block at a time while it runs.</span>
pipe.transformer_ref.enable_group_offload(
    onload_device=torch.device(<span class="hljs-string">&quot;cuda&quot;</span>),
    offload_device=torch.device(<span class="hljs-string">&quot;cpu&quot;</span>),
    offload_type=<span class="hljs-string">&quot;block_level&quot;</span>,
    num_blocks_per_group=<span class="hljs-number">1</span>,
    use_stream=<span class="hljs-literal">True</span>,
)

results = pipe(
    prompt=<span class="hljs-string">&quot;subject_definitions: &lt;Subject 1&gt; is the woman in &lt;Picture 1&gt;. summary: [reference generation] &lt;Subject 1&gt; waves at the camera.&quot;</span>,
    references=[
        PikaH3FlexImageReference.from_file(<span class="hljs-string">&quot;subject.png&quot;</span>),
        PikaH3FlexVideoReference.from_file(<span class="hljs-string">&quot;motion_ref.mp4&quot;</span>),  <span class="hljs-comment"># frames, their `fps` and the soundtrack</span>
        PikaH3FlexAudioReference.from_file(<span class="hljs-string">&quot;voice.wav&quot;</span>),  <span class="hljs-comment"># waveform and its `sample_rate`</span>
    ],
    height=<span class="hljs-number">768</span>,
    width=<span class="hljs-number">1344</span>,
    num_frames=<span class="hljs-number">124</span>,
    generator=torch.Generator(<span class="hljs-string">&quot;cuda&quot;</span>).manual_seed(<span class="hljs-number">42</span>),
    output=[<span class="hljs-string">&quot;videos&quot;</span>, <span class="hljs-string">&quot;audio&quot;</span>, <span class="hljs-string">&quot;sampling_rate&quot;</span>],
)<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="diffusers.PikaH3FlexModularPipeline" 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.PikaH3FlexModularPipeline"><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><!---->PikaH3FlexModularPipeline<!----></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.PikaH3FlexModularPipeline"><!----><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">sınıf</span> <span class="font-medium">diffusers.</span><span class="font-semibold">PikaH3FlexModularPipeline</span></span></h3><!----> <a id="diffusers.PikaH3FlexModularPipeline" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.PikaH3FlexModularPipeline"><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/pika_h3_flex/modular_pipeline.py#L211" 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>Pika H3 Flex ile ortak video + ses üretimi için bir ModularPipeline, post-trained altı adımlı MiniMax-H3: <code>t2va</code> (yalnızca metin) ve <code>fl2va</code> (ilk ve/veya son ana kare) iş akışları <code>transformer/</code> checkpoint bölümüne karşı, <code>fl2va</code> adaptörü ile; ve <code>ref2va</code> (omni-reference) iş akışı <code>transformer_ref/</code> bölümüne karşı, <code>ref2va</code> adaptörü ile.</p> <p>Temel bileşenler MiniMax-H3’ünkilerdir ve Pika H3 Flex sürümü, iki transformer bölümünün üzerine yerleştirilen bir çift rank-128 LoRA adaptörüdür; bu nedenle bir pipeline iki adımda oluşturulur: bileşenler, temel bileşenleri ve iki zamanlayıcıyı adlandıran bir <code>modular_model_index.json</code> üzerinden yüklenir ve seçilen iş akışının adaptörü <a href="/docs/diffusers/main/tr/api/loaders/lora#diffusers.loaders.MiniMaxH3LoraLoaderMixin.load_lora_weights">load_lora_weights()</a> ile yüklenir.</p> <div class="relative group rounded-md"><a id="diffusers.PikaH3FlexModularPipeline.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.PikaH3FlexModularPipeline.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> Kopyalandı</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;Pika-Labs/Pika-H3-Flex&quot;</span>, workflow=<span class="hljs-string">&quot;fl2va&quot;</span>)
pipe.load_components(dtype={<span class="hljs-string">&quot;vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;audio_vae&quot;</span>: torch.float32, <span class="hljs-string">&quot;default&quot;</span>: torch.bfloat16})
pipe.load_lora_weights(<span class="hljs-string">&quot;Pika-Labs/Pika-H3-Flex&quot;</span>, weight_name=<span class="hljs-string">&quot;fl2va/adapter.safetensors&quot;</span>)<!----></pre></div><!----></div><!----> <p>MiniMax-H3 gibi, Pika H3 Flex de metin koşullandırma, anahtar kare veya referans koşullandırma latentleri, ses latentleri ve video latentlerini aynı anda barındıran <strong>tek bir paketlenmiş diziyi</strong> gürültüden arındırır. Bu nedenle bloklar, modalite başına tensörler yerine bir satır düzenini (row layout) devreder ve pipeline, tek bir transformer çağrısı içinde adım atılan iki zamanlayıcıyı (<code>shift = 12.0</code> video için, <code>shift = 3.0</code> ses için) taşır. MiniMax-H3'ten farklı olarak, bu iki zamanlayıcı <code>denoising_step_list</code> sabit altı adımlık listesi üzerinde <code>stochastic_sampling</code> kullanan <a href="/docs/diffusers/main/tr/api/schedulers/minimax_h3#diffusers.MiniMaxH3Scheduler">MiniMaxH3Scheduler</a> örnekleridir: <code>num_inference_steps</code> yoktur ve her adım, isteğin üretecinden (generator) taze gürültü çeker.</p> <p>Uyarlama katmanları, taban modelleri gibi rehberlik damıtılmıştır: bir rehber, <code>negative_prompt</code> veya <code>guidance_scale</code> yoktur ve her adım tam olarak bir ileri geçiş çalıştırır.</p> <p>Bu modül, her bloğun dayandığı model bilgilerini taşır — yapılandırma türevli geometri özellikler olarak, tuval, katman ve kare sayısı aritmetiği işlevler olarak — ve koşullandırma, kodlama ve gürültü sözleşmeleri blokların kendisinde yaşar.</p></div> <!--[1--><h2 class="relative group"><a id="diffusers.PikaH3FlexBlocks" 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.PikaH3FlexBlocks"><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><!---->PikaH3FlexBlocks<!----></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.PikaH3FlexBlocks"><!----><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">sınıf</span> <span class="font-medium">diffusers.</span><span class="font-semibold">PikaH3FlexBlocks</span></span></h3><!----> <a id="diffusers.PikaH3FlexBlocks" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.PikaH3FlexBlocks"><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/pika_h3_flex/modular_blocks_pika_h3_flex.py#L668" 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>Pika H3 Flex ile ortak video + ses üretimi için Auto Modular pipeline blokları: <code>references</code> ve anahtar kare girdilerine göre seçilen <code>t2va</code> (yalnızca metin), <code>fl2va</code> (ilk ve/veya son anahtar kare) ve <code>ref2va</code> (omni-referans) iş akışları. <code>t2va</code> ve <code>fl2va</code> iş akışları <code>transformer</code> bölümünü <code>fl2va</code> adaptörüyle, <code>ref2va</code> ise <code>transformer_ref</code> bölümünü <code>ref2va</code> adaptörüyle çalıştırır; bu nedenle isteğin gittiği iş akışının adaptörünü yükleyin. <code>workflow=</code> verilmediğinde, bileşenlerin yüklenmesi <strong>her iki</strong> 61,7 GB’lık transformer bölümünü de çeker — görevin gerektirdiği şeyi yüklemek için birini belirtin.</p> <p>Desteklenen iş akışları:</p> <ul><li><code>t2va</code>: <code>prompt</code> gerektirir</li> <li><code>fl2va</code>: <code>prompt</code>, <code>image</code> veya <code>prompt</code>, <code>last_image</code> gerektirir</li> <li><code>ref2va</code>: <code>prompt</code> ve <code>references</code> gerektirir</li></ul> <p>Bileşenler:
image_processor (<code>VaeImageProcessor</code>) text_encoder (<code>Qwen3VLForConditionalGeneration</code>) tokenizer
(<code>Qwen2Tokenizer</code>) processor (<code>Qwen3VLProcessor</code>) vae (<code>AutoencoderKLMiniMaxH3</code>) audio_vae
(<code>AutoencoderKLMiniMaxH3Audio</code>) scheduler (<code>MiniMaxH3Scheduler</code>) audio_scheduler (<code>MiniMaxH3Scheduler</code>)
transformer_ref (<code>MiniMaxH3Transformer3DModel</code>) transformer (<code>MiniMaxH3Transformer3DModel</code>) video_processor
(<code>VideoProcessor</code>)</p> <p>Yapılandırmalar:
canvas_short_edge (varsayılan: 768) canvas_max_pixels (varsayılan: 1032192) denoising_step_list (varsayılan: [1000,
750, 500, 250, 100, 40, 0])</p> <p>Girdiler:
references (<code>list</code>, <em>isteğe bağlı</em>):
Koşullandırılacak referanslar, <strong>modelin onları okuması gereken sırayla</strong>: sıra, bunları istek sunumunda etiketler ve paylaşımlı döner saat üzerinde yerleştirir, bu nedenle farklı bir sıra farklı bir istektir. Her modallite için bir dataclass, hepsi bellek içi medya tutar — en fazla 9 adet <code>PikaH3FlexImageReference</code>, kendi <code>fps</code> değerinde en fazla 3 adet <code>PikaH3FlexVideoReference</code> (kendi <code>audio</code> ses bandı da koşullandırılır) veya kendi <code>sample_rate</code> değerinde en fazla 3 adet <code>PikaH3FlexAudioReference</code> — toplamda en fazla 12 referans ve ses referansları tek başına olamaz. Her video veya bağımsız ses referansı 2 ila 15 saniye sürer ve video süreleri ile bağımsız ses süreleri her biri en fazla 15 saniyeye kadar toplanır. Bu bloklar asla bir medya dosyası açmaz: hızları da getiren her sınıfın <code>from_file</code> sınıf metoduyla çözümlenir.
height (<code>int</code>, <em>isteğe bağlı</em>):
Üretilen videonun piksel cinsinden yüksekliği, 32'nin bir katı.
width (<code>int</code>, <em>isteğe bağlı</em>):
Üretilen videonun piksel cinsinden genişliği, 32'nin bir katı.
num_frames (<code>int</code>, <em>isteğe bağlı</em>):
Sabit 24 fps'de üretilecek kare sayısı: 120 ila 360, yani 5 ila 15 saniye, video VAE'nin çözebileceği bir sonraki <code>17 * n + 5</code> değerine yuvarlanır. Referans ses bandı kadar uzun bir video üretmek için <code>round(samples / sample_rate * 24)</code> geçirin.
image (<code>Image</code>, <em>isteğe bağlı</em>):
Videonun başladığı anahtar kare. Tuvalin en-boy oranına göre ortadan kırpılır ve üzerine yeniden boyutlandırılır; tuval varsayılan olarak kendi en-boy oranından türetilir.
last_image (<code>Image</code>, <em>isteğe bağlı</em>):
Videonun bittiği anahtar kare. Bir kareye <em>kadar</em> üretmek için tek başına geçirilebilir. <code>image</code> gibi, tuvalin en-boy oranına göre ortadan kırpılır ve üzerine yeniden boyutlandırılır.
prompt (<code>str</code>):
Üretimi yönlendirmek için istek, tek bir dize.
normalized_references (<code>list</code>, <em>isteğe bağlı</em>):
Kurulum adımı tarafından normalize edilen referanslar, paketlenmiş sırayla.
keyframes (<code>list</code>, <em>isteğe bağlı</em>):
Hedef tuvale yerleştirilen anahtar kareler, paketlenmiş sırayla.
condition_latents (<code>list</code>, <em>isteğe bağlı</em>):
Kodlanmış video koşullandırma latentleri, paketlenmiş sırayla her görüntü ve video referansı için bir adet. Şekilleri, her referans bloğunun geometrisinin geldiği yerdir.
audio_condition_latents (<code>list</code>, <em>isteğe bağlı</em>):
Kodlanmış ses koşullandırma satırları, paketlenmiş sırayla ses içeren her referans için bir adet.
generator (<code>Generator</code>, <em>isteğe bağlı</em>):
İsteğin üreteci. Koşullandırma gürültüsü, üretilen satırların gürültüsünden önce, her koşul için bir çekimle buradan çekilir.
latents (<code>Tensor</code>):
<code>(1, 24, num_latent_frames, latent_height, latent_width)</code> şeklinde önceden üretilmiş video gürültüsü, çekim yerine kullanılır.
audio_latents (<code>Tensor</code>):
<code>(2, 32, num_audio_latents)</code> şeklinde önceden üretilmiş ses gürültüsü.
*<em>denoiser_input_fields (<code>None</code>,</em>isteğe bağlı<em>):
Transformer'ın adla okuduğu paketlenmiş dizinin yapısal tanımı: <code>token_tags</code>, <code>position_ids</code> ve üç satır-dizin tensörü.
attention_kwargs (<code>dict</code>,</em>isteğe bağlı<em>):
Dikkat işleyicileri için ek kwargs.
keyframe_anchors (<code>tuple</code>,</em>isteğe bağlı<em>, varsayılan ()):
Her anahtar karenin videonun hangi ucuna sabitlendiği, paketlenmiş sırayla.
output_type (<code>str</code>,</em>isteğe bağlı*, varsayılan pil):
Çıktı biçimi: ‘pil’, ‘np’ veya ‘pt’.</p> <p>Çıktılar:
videos (<code>list</code>):
Oluşturulan video.
audio (<code>Tensor</code>):
<code>(1, 2, num_samples)</code> biçiminde oluşturulan ses bandı.
sampling_rate (<code>int</code>):
Hz cinsinden oluşturulan ses bandının örneklem hızı.</p></div> <!--[1--><h2 class="relative group"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference" 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.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference"><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><!---->PikaH3FlexImageReference<!----></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.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference"><!----><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">sınıf</span> <span class="font-medium">diffusers.modular_pipelines.pika_h3_flex.</span><span class="font-semibold">PikaH3FlexImageReference</span></span></h3><!----> <a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference"><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/pika_h3_flex/references.py#L83" 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-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">image<span class="opacity-60">: typing.Union[PIL.Image.Image, numpy.ndarray, torch.Tensor]</span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[0--><p class="flex items-center font-semibold !mt-2 !mb-2 text-gray-800">Parametreler <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700 ml-3"></span></p> <ul class="px-2"><!--[--><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.image" 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.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.image"><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> <span><!----><strong>image</strong> (<code>PIL.Image.Image</code>, <code>np.ndarray</code> veya <code>torch.Tensor</code>) —
Referans görüntü: bir görüntü, <code>(height, width, 3)</code> boyutunda bir dizi veya <code>(3, height, width)</code> boyutunda bir tensör, <code>uint8</code> veya
<code>[0, 1]</code> aralığında kayan noktalı. Üretilen geometriyi asla bağlamaz; kendi kısa kenarında kodlanır, yayınlanan checkpoint için 2048, istek hangi tuvalde üretirse üretsin.<!----></span></span></li><!--]--></ul><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Bir konu, stil veya sahne referansı: istek başına en fazla 9.</p> <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.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.from_file"><!----><h4 class="!m-0"><span class="flex-1 rounded-xl py-0.5 break-all bg-gradient-to-r from-blue-50/60 to-white dark:from-gray-900 dark:to-gray-950 text-blue-700 dark:text-blue-300 font-medium px-2"><svg width="1em" height="1em" viewBox="0 0 32 33" class="mr-1 inline-block -mt-0.5"  xmlns="http://www.w3.org/2000/svg"><path d="M5.80566 18.3545C4.90766 17.4565 4.90766 16.0005 5.80566 15.1025L14.3768 6.53142C15.2748 5.63342 16.7307 5.63342 17.6287 6.53142L26.1999 15.1025C27.0979 16.0005 27.0979 17.4565 26.1999 18.3545L17.6287 26.9256C16.7307 27.8236 15.2748 27.8236 14.3768 26.9256L5.80566 18.3545Z" fill="currentColor" fill-opacity="0.25"/><path fill-rule="evenodd" clip-rule="evenodd" d="M16.4801 13.9619C16.4801 12.9761 16.7467 12.5436 16.9443 12.3296C17.1764 12.078 17.5731 11.8517 18.2275 11.707C18.8821 11.5623 19.638 11.5342 20.4038 11.5582C20.7804 11.57 21.1341 11.5932 21.4719 11.6156L21.5263 11.6193C21.8195 11.6389 22.1626 11.6618 22.4429 11.6618V7.40825C22.3209 7.40825 22.1219 7.39596 21.7544 7.37149C21.4202 7.34925 20.9976 7.32115 20.5371 7.30672C19.6286 7.27824 18.4672 7.29779 17.3093 7.55377C16.1512 7.8098 14.8404 8.33724 13.8181 9.4452C12.7612 10.5907 12.2266 12.1236 12.2266 13.9619V15.0127H10.6836V19.2662H12.2266V26.6332H16.4801V19.2662H20.3394V15.0127H16.4801V13.9619Z" fill="currentColor"/></svg>from_file</span></h4><!----> <a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.from_file" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.from_file"><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/pika_h3_flex/references.py#L99" 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-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">media<span class="opacity-60"></span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[0--><p class="flex items-center font-semibold !mt-2 !mb-2 text-gray-800">Parametreler <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700 ml-3"></span></p> <ul class="px-2"><!--[--><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.from_file.media" 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.modular_pipelines.pika_h3_flex.PikaH3FlexImageReference.from_file.media"><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> <span><!----><strong>media</strong> (<code>str</code> veya <code>os.PathLike</code>) — Görselin yolu veya URL'si.<!----></span></span></li><!--]--></ul><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Bir görüntü dosyasını <a href="/docs/diffusers/main/tr/api/utilities#diffusers.utils.load_image">load_image()</a> üzerinden <code>PikaH3FlexImageReference</code> içine yükleyin.</p></div></div> <!--[1--><h2 class="relative group"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference" 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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference"><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><!---->PikaH3FlexVideoReference<!----></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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference"><!----><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">sınıf</span> <span class="font-medium">diffusers.modular_pipelines.pika_h3_flex.</span><span class="font-semibold">PikaH3FlexVideoReference</span></span></h3><!----> <a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference"><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/pika_h3_flex/references.py#L111" 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-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">frames<span class="opacity-60">: typing.Union[list[PIL.Image.Image], numpy.ndarray, torch.Tensor]</span></span></span><span class="comma cursor-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">fps<span class="opacity-60">: float | None = None</span></span></span><span class="comma cursor-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">audio<span class="opacity-60">: typing.Optional[torch.Tensor] = None</span></span></span><span class="comma cursor-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">sample_rate<span class="opacity-60">: int | None = None</span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[0--><p class="flex items-center font-semibold !mt-2 !mb-2 text-gray-800">Parametreler <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700 ml-3"></span></p> <ul class="px-2"><!--[--><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.frames" 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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.frames"><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> <span><!----><strong>frames</strong> (<code>list[PIL.Image.Image]</code>, <code>np.ndarray</code> veya <code>torch.Tensor</code>) —
Referans kareler: bir görüntü listesi, <code>(num_frames, height, width, 3)</code> boyutunda bir dizi veya <code>(num_frames, 3, height, width)</code> boyutunda bir tensör, <code>uint8</code> veya <code>[0, 1]</code> aralığında kayan noktalı.<!----></span></span></li><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.fps" 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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.fps"><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> <span><!----><strong>fps</strong> (<code>float</code>, <em>isteğe bağlı</em>, varsayılan 24.0) —
<code>frames</code>’in taşıdığı kare hızı; referansın görme bloklarını conditioner’ın 2 fps ızgarasına yerleştiren değer budur. MiniMax-H3’ün kendi saati 24 fps olduğundan, başka bir hız tüm kareleri atlayarak ve çoğaltarak bu hıza yeniden örneklenir; bu da kareler bir dosyadan geldiğinde doğru ayarlanacak alanı oluşturur.<!----></span></span></li><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.audio" 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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.audio"><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> <span><!----><strong>audio</strong> (<code>torch.Tensor</code>, şekli <code>(channels, num_samples)</code>, <em>isteğe bağlı</em>) —
Bu videonun ses bandı, mono veya stereo; kendi referansı olarak değil, referansın kendi ses bandı olarak koşullandırılır. Bu parametre verilmezse, referans yalnızca harekete göre koşullandırılır.<!----></span></span></li><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.sample_rate" 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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.sample_rate"><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> <span><!----><strong>sample_rate</strong> (<code>int</code>, <em>isteğe bağlı</em>) —
<code>audio</code> örneklerini taşıdığı hız. Belirtilmezse, ses VAE'sinin kendi hızı kullanılır ve örnekler olduğu gibi bırakılır; başka bir hız belirtilirse örnekler bu hıza yeniden örneklenir.<!----></span></span></li><!--]--></ul><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Bir hareket ve kamera referansı: istek başına en fazla 3 adet, kendi ses efektleriyle birlikte koşullandırılır.</p> <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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.from_file"><!----><h4 class="!m-0"><span class="flex-1 rounded-xl py-0.5 break-all bg-gradient-to-r from-blue-50/60 to-white dark:from-gray-900 dark:to-gray-950 text-blue-700 dark:text-blue-300 font-medium px-2"><svg width="1em" height="1em" viewBox="0 0 32 33" class="mr-1 inline-block -mt-0.5"  xmlns="http://www.w3.org/2000/svg"><path d="M5.80566 18.3545C4.90766 17.4565 4.90766 16.0005 5.80566 15.1025L14.3768 6.53142C15.2748 5.63342 16.7307 5.63342 17.6287 6.53142L26.1999 15.1025C27.0979 16.0005 27.0979 17.4565 26.1999 18.3545L17.6287 26.9256C16.7307 27.8236 15.2748 27.8236 14.3768 26.9256L5.80566 18.3545Z" fill="currentColor" fill-opacity="0.25"/><path fill-rule="evenodd" clip-rule="evenodd" d="M16.4801 13.9619C16.4801 12.9761 16.7467 12.5436 16.9443 12.3296C17.1764 12.078 17.5731 11.8517 18.2275 11.707C18.8821 11.5623 19.638 11.5342 20.4038 11.5582C20.7804 11.57 21.1341 11.5932 21.4719 11.6156L21.5263 11.6193C21.8195 11.6389 22.1626 11.6618 22.4429 11.6618V7.40825C22.3209 7.40825 22.1219 7.39596 21.7544 7.37149C21.4202 7.34925 20.9976 7.32115 20.5371 7.30672C19.6286 7.27824 18.4672 7.29779 17.3093 7.55377C16.1512 7.8098 14.8404 8.33724 13.8181 9.4452C12.7612 10.5907 12.2266 12.1236 12.2266 13.9619V15.0127H10.6836V19.2662H12.2266V26.6332H16.4801V19.2662H20.3394V15.0127H16.4801V13.9619Z" fill="currentColor"/></svg>from_file</span></h4><!----> <a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.from_file" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.from_file"><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/pika_h3_flex/references.py#L147" 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-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">media<span class="opacity-60"></span></span></span><!--]--> <span>)</span> <!--[0--><span class="font-bold">→</span> <span class="rounded hover:bg-gray-400 cursor-pointer"><!----><span><code>PikaH3FlexVideoReference</code></span><!----></span><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[0--><p class="flex items-center font-semibold !mt-2 !mb-2 text-gray-800">Parametreler <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700 ml-3"></span></p> <ul class="px-2"><!--[--><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.from_file.media" 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.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.from_file.media"><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> <span><!----><strong>media</strong> (<code>str</code> veya <code>os.PathLike</code>) — Videoyun yolu veya URL'si.<!----></span></span></li><!--]--></ul><!--]--> <!--[-1--><!--]--> <!--[0--><div id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexVideoReference.from_file.returns" class="flex items-center font-semibold space-x-3 text-base !mt-0 !mb-0 text-gray-800 rounded "><p class="text-base"><!---->Dönüş değeri<!----></p> <!--[0--><!---->


<p><code>PikaH3FlexVideoReference</code></p>
<!----><!--]--> <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700"></span></div> <div class="text-base"><!---->


<p>Kapsayıcının bildirdiği kare hızında, ses akışı varsa ses kaydını ve o ses kaydının kendi örneklem hızını taşıyan <code>(num_frames, height, width, 3)</code> <code>uint8</code> kareler.</p>
<!----></div><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Bir video dosyasını, çözünürlüğünde, kare hızında ve taşıdığı ses kaydında <code>PikaH3FlexVideoReference</code> olarak çözümlen.</p> <p>Oranlar referansa yerleşir; bu, <a href="/docs/diffusers/main/tr/api/utilities#diffusers.utils.load_video">load_video()</a> yerine bu şekilde çözümleme yapmanın amacıdır: MiniMax-H3, referansı kendi 24 fps'ine yeniden örnekler, bu yüzden giriş sırasında kaybedilen bir kare hızı, yanlış hızda koşullandırılmış bir istemektir ve bunu bildirecek hiçbir şey yoktur. Meta verileri yanlış olan bir kapsayıcı, döndürülen referans üzerinde <code>fps</code> veya <code>sample_rate</code> değerlerini geçersiz kılarak düzeltilir.</p> <p><a href="https://github.com/PyAV-Org/PyAV" rel="nofollow">PyAV</a> gerektirir.</p></div></div> <!--[1--><h2 class="relative group"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference" 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.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference"><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><!---->PikaH3FlexAudioReference<!----></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.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference"><!----><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">sınıf</span> <span class="font-medium">diffusers.modular_pipelines.pika_h3_flex.</span><span class="font-semibold">PikaH3FlexAudioReference</span></span></h3><!----> <a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference"><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/pika_h3_flex/references.py#L174" 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-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">audio<span class="opacity-60">: Tensor</span></span></span><span class="comma cursor-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">sample_rate<span class="opacity-60">: int | None = None</span></span></span><!--]--> <span>)</span> <!--[-1--><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[0--><p class="flex items-center font-semibold !mt-2 !mb-2 text-gray-800">Parametreler <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700 ml-3"></span></p> <ul class="px-2"><!--[--><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.audio" 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.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.audio"><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> <span><!----><strong>audio</strong> (<code>torch.Tensor</code>, şekli <code>(channels, num_samples)</code>) —
Referans dalga formu, mono veya stereo.<!----></span></span></li><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.sample_rate" 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.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.sample_rate"><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> <span><!----><strong>sample_rate</strong> (<code>int</code>, <em>isteğe bağlı</em>) —
<code>audio</code> örneklerini taşıdığı hız. Belirtilmezse, ses VAE'sinin kendi hızı kullanılır ve örnekler olduğu gibi bırakılır; başka bir hız belirtilirse örnekler bu hıza yeniden örneklenir.<!----></span></span></li><!--]--></ul><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Ses veya müzik referansı: istek başına en fazla 3 adet ve asla tek başına değil — bir ses referansı, en az bir görsel veya video referansı ile eşleştirilmelidir. Hiçbir zaman conditioner'a ulaşmaz ve yalnızca ses VAE tarafından kodlanır.</p> <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.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.from_file"><!----><h4 class="!m-0"><span class="flex-1 rounded-xl py-0.5 break-all bg-gradient-to-r from-blue-50/60 to-white dark:from-gray-900 dark:to-gray-950 text-blue-700 dark:text-blue-300 font-medium px-2"><svg width="1em" height="1em" viewBox="0 0 32 33" class="mr-1 inline-block -mt-0.5"  xmlns="http://www.w3.org/2000/svg"><path d="M5.80566 18.3545C4.90766 17.4565 4.90766 16.0005 5.80566 15.1025L14.3768 6.53142C15.2748 5.63342 16.7307 5.63342 17.6287 6.53142L26.1999 15.1025C27.0979 16.0005 27.0979 17.4565 26.1999 18.3545L17.6287 26.9256C16.7307 27.8236 15.2748 27.8236 14.3768 26.9256L5.80566 18.3545Z" fill="currentColor" fill-opacity="0.25"/><path fill-rule="evenodd" clip-rule="evenodd" d="M16.4801 13.9619C16.4801 12.9761 16.7467 12.5436 16.9443 12.3296C17.1764 12.078 17.5731 11.8517 18.2275 11.707C18.8821 11.5623 19.638 11.5342 20.4038 11.5582C20.7804 11.57 21.1341 11.5932 21.4719 11.6156L21.5263 11.6193C21.8195 11.6389 22.1626 11.6618 22.4429 11.6618V7.40825C22.3209 7.40825 22.1219 7.39596 21.7544 7.37149C21.4202 7.34925 20.9976 7.32115 20.5371 7.30672C19.6286 7.27824 18.4672 7.29779 17.3093 7.55377C16.1512 7.8098 14.8404 8.33724 13.8181 9.4452C12.7612 10.5907 12.2266 12.1236 12.2266 13.9619V15.0127H10.6836V19.2662H12.2266V26.6332H16.4801V19.2662H20.3394V15.0127H16.4801V13.9619Z" fill="currentColor"/></svg>from_file</span></h4><!----> <a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.from_file" class="header-link invisible with-hover:group-hover:visible pr-2" href="#diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.from_file"><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/pika_h3_flex/references.py#L193" 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-pointer"><span class="rounded hover:bg-black hover:text-white dark:hover:bg-white dark:hover:text-black">media<span class="opacity-60"></span></span></span><!--]--> <span>)</span> <!--[0--><span class="font-bold">→</span> <span class="rounded hover:bg-gray-400 cursor-pointer"><!----><span><code>PikaH3FlexAudioReference</code></span><!----></span><!--]--></p><!--]--> <div class="!mb-10 relative docstring-details "><!--[-1--><!--]--> <!--[0--><p class="flex items-center font-semibold !mt-2 !mb-2 text-gray-800">Parametreler <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700 ml-3"></span></p> <ul class="px-2"><!--[--><li class="text-base !pl-4 my-3 rounded "><span class="group flex space-x-1.5 items-start"><a id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.from_file.media" 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.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.from_file.media"><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> <span><!----><strong>media</strong> (<code>str</code> veya <code>os.PathLike</code>) — Sesin veya ses bandı alınacak videonun yolu veya URL'si.<!----></span></span></li><!--]--></ul><!--]--> <!--[-1--><!--]--> <!--[0--><div id="diffusers.modular_pipelines.pika_h3_flex.PikaH3FlexAudioReference.from_file.returns" class="flex items-center font-semibold space-x-3 text-base !mt-0 !mb-0 text-gray-800 rounded "><p class="text-base"><!---->Dönüş değeri<!----></p> <!--[0--><!---->


<p><code>PikaH3FlexAudioReference</code></p>
<!----><!--]--> <span class="flex-auto border-t-2 border-gray-100 dark:border-gray-700"></span></div> <div class="text-base"><!---->


<p>kap konteynerinin bildirdiği örnek hızında <code>(channels, num_samples)</code> boyutlu float32 dalga formu.</p>
<!----></div><!--]--> <!--[-1--><!--]--></div></div><!----> <p>Bir ses dosyasını, taşıdığı örneklem hızında <code>PikaH3FlexAudioReference</code> olarak çözümlen.</p> <p><a href="https://github.com/PyAV-Org/PyAV" rel="nofollow">PyAV</a> gerektirir.</p></div></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/pika_h3_flex.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_1rjuniz = {
						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.vZc-dOUR.js"),
						import("/docs/diffusers/main/en/_app/immutable/entry/app.B6c3Ml05.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 223],
							data: [null,null],
							form: null,
							error: null
						});
					});
				}
			</script>
		
Xet Storage Details

Size:: 109 kB
Xet hash:: ca98be03d86023ca7813902e35444e9440320509d56c6e52bd62998e072fe02f
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.