transformers/pr_26617/en/expert_parallelism.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Expert parallelism&quot;,&quot;local&quot;:&quot;expert-parallelism&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;DistributedConfig&quot;,&quot;local&quot;:&quot;distributedconfig&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}">
		<link href="/docs/transformers/pr_26617/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/entry/start.b5ae2c21.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/scheduler.31fdf58d.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/singletons.512cdb48.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/index.252883d5.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/paths.81255c3b.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/entry/app.9acf2c3e.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/preload-helper.bb442aeb.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/index.2f76fdf0.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/nodes/0.da6b3909.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/each.e59479a4.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/nodes/38.8c0f98aa.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/CopyLLMTxtMenu.a69e059a.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/MermaidChart.svelte_svelte_type_style_lang.e4c7f916.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/IconCopy.ac192424.js">
		<link rel="modulepreload" href="/docs/transformers/pr_26617/en/_app/immutable/chunks/CodeBlock.ab12f8e1.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Expert parallelism&quot;,&quot;local&quot;:&quot;expert-parallelism&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;DistributedConfig&quot;,&quot;local&quot;:&quot;distributedconfig&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END -->      <p></p>   <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div 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>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> </div>  <h1 class="relative group"><a id="expert-parallelism" 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="#expert-parallelism"><span><svg class="" 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>Expert parallelism</span></h1> <p data-svelte-h="svelte-1sczd5z"><a href="https://huggingface.co/spaces/nanotron/ultrascale-playbook?section=expert_parallelism" rel="nofollow">Expert parallelism</a> is a parallelism strategy for <a href="https://huggingface.co/blog/moe" rel="nofollow">mixture-of-experts (MoE) models</a>. Each expert’s feedforward layer lives on a different hardware accelerator. A router dispatches tokens to the appropriate experts and gathers the results. This approach scales models to far larger parameter counts without increasing computation cost because each token activates only a few experts.</p>  <h2 class="relative group"><a id="distributedconfig" 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="#distributedconfig"><span><svg class="" 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>DistributedConfig</span></h2> <blockquote class="warning" data-svelte-h="svelte-wrs6j"><p>The <code>DistributedConfig</code> API is experimental and its usage may change in the future.</p></blockquote> <p data-svelte-h="svelte-1v9ynxu">Enable expert parallelism with the <code>DistributedConfig</code> class and the <code>enable_expert_parallel</code> argument.</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 class="" 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> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoModelForCausalLM, AutoTokenizer
<span class="hljs-keyword">from</span> transformers.distributed.configuration_utils <span class="hljs-keyword">import</span> DistributedConfig

distributed_config = DistributedConfig(enable_expert_parallel=<span class="hljs-literal">True</span>)

model = AutoModelForCausalLM.from_pretrained(
    <span class="hljs-string">&quot;openai/gpt-oss-120b&quot;</span>,
    dtype=<span class="hljs-string">&quot;auto&quot;</span>,
    distributed_config=distributed_config,
)<!-- HTML_TAG_END --></pre></div> <blockquote class="tip" data-svelte-h="svelte-1j6as4h"><p>Expert parallelism automatically enables <a href="./perf_infer_gpu_multi">tensor parallelism</a> for attention layers.</p></blockquote> <p data-svelte-h="svelte-p14elo">This argument switches to the <code>ep_plan</code> (expert parallel plan) defined in each MoE model’s config file. The <code>GroupedGemmParallel</code> class splits expert weights so each device loads only its local experts. The <code>ep_router</code> routes tokens to experts and an all-reduce operation combines their outputs.</p> <p data-svelte-h="svelte-ep1vv0">Launch your inference script with <a href="https://pytorch.org/docs/stable/elastic/run.html" rel="nofollow">torchrun</a> and specify how many devices to use. The number of devices must evenly divide the total number of experts.</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 class="" 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> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->torchrun --nproc-per-<span class="hljs-keyword">node</span> <span class="hljs-title">8</span> your_script.py<!-- HTML_TAG_END --></pre></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/transformers/blob/main/docs/source/en/expert_parallelism.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 data-svelte-h="svelte-zjs2n5"><span class="underline">Update</span> on GitHub</span></a>  <p></p> 
			
			<script>
				{
					__sveltekit_1x0t0ja = {
						assets: "/docs/transformers/pr_26617/en",
						base: "/docs/transformers/pr_26617/en",
						env: {}
					};

					const element = document.currentScript.parentElement;

					const data = [null,null];

					Promise.all([
						import("/docs/transformers/pr_26617/en/_app/immutable/entry/start.b5ae2c21.js"),
						import("/docs/transformers/pr_26617/en/_app/immutable/entry/app.9acf2c3e.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 38],
							data,
							form: null,
							error: null
						});
					});
				}
			</script>