transformers/pr_36839/en/perf_infer_gpu_multi.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Distributed GPU inference&quot;,&quot;local&quot;:&quot;distributed-gpu-inference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}">
		<link href="/docs/transformers/pr_36839/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/entry/start.6be8d590.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/scheduler.01eeda35.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/singletons.177df05e.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/index.4862150a.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/paths.517376d1.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/entry/app.09748b4b.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/index.6dd51b66.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/nodes/0.8897c14d.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/each.e59479a4.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/nodes/415.b0559481.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/Tip.de9bae2b.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/CodeBlock.864da1b0.js">
		<link rel="modulepreload" href="/docs/transformers/pr_36839/en/_app/immutable/chunks/EditOnGithub.7faefd25.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Distributed GPU inference&quot;,&quot;local&quot;:&quot;distributed-gpu-inference&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END -->      <p></p>   <h1 class="relative group"><a id="distributed-gpu-inference" 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="#distributed-gpu-inference"><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>Distributed GPU inference</span></h1> <p data-svelte-h="svelte-hxh0ev"><a href="./perf_train_gpu_many#tensor-parallelism">Tensor parallelism</a> shards a model onto multiple GPUs and parallelizes computations such as matrix multiplication. It enables fitting larger model sizes into memory and is faster because each GPU can process a tensor slice.</p>  <div class="course-tip  bg-gradient-to-br dark:bg-gradient-to-r before:border-green-500 dark:before:border-green-800 from-green-50 dark:from-gray-900 to-white dark:to-gray-950 border border-green-50 text-green-700 dark:text-gray-400"><p data-svelte-h="svelte-1awp5l7">Expand the list below to see which models support tensor parallelism. Open a GitHub issue or pull request to add support for a model not currently below.</p></div> <details data-svelte-h="svelte-svh7x"><summary>Supported models</summary> <ul><li><a href="./model_doc/cohere">Cohere</a> and <a href="./model_doc/cohere2">Cohere 2</a></li> <li><a href="./model_doc/gemma">Gemma</a> and <a href="./model_doc/gemma2">Gemma 2</a></li> <li><a href="./model_doc/glm">GLM</a></li> <li><a href="./model_doc/granite">Granite</a></li> <li><a href="./model_doc/llama">Llama</a></li> <li><a href="./model_doc/mistral">Mistral</a></li> <li><a href="./model_doc/mixtral">Mixtral</a></li> <li><a href="./model_doc/olmo">OLMo</a> and <a href="./model_doc/olmo2">OLMo2</a></li> <li><a href="./model_doc/phi">Phi</a> and <a href="./model_doc/phi3">Phi-3</a></li> <li><a href="./model_doc/qwen2">Qwen2</a>, <a href="./model_doc/qwen2_moe">Qwen2Moe</a>, and <a href="./model_doc/qwen2_5_vl">Qwen2-VL</a></li> <li><a href="./model_doc/starcoder2">Starcoder2</a></li></ul></details> <p data-svelte-h="svelte-t4pegj">Set <code>tp_plan=&quot;auto&quot;</code> in <a href="/docs/transformers/pr_36839/en/model_doc/auto#transformers.AutoModel.from_pretrained">from_pretrained()</a> to enable tensor parallelism for inference.</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> os
<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-comment"># initialize distributed environment</span>
rank = <span class="hljs-built_in">int</span>(os.environ[<span class="hljs-string">&quot;RANK&quot;</span>])
device = torch.device(<span class="hljs-string">f&quot;cuda:<span class="hljs-subst">{rank}</span>&quot;</span>)
torch.cuda.set_device(device)
torch.distributed.init_process_group(<span class="hljs-string">&quot;nccl&quot;</span>, device_id=device)

<span class="hljs-comment"># enable tensor parallelism</span>
model = AutoModelForCausalLM.from_pretrained(
    <span class="hljs-string">&quot;meta-llama/Meta-Llama-3-8B-Instruct&quot;</span>,
    tp_plan=<span class="hljs-string">&quot;auto&quot;</span>,
)

<span class="hljs-comment"># prepare input tokens</span>
tokenizer = AutoTokenizer.from_pretrained(<span class="hljs-string">&quot;meta-llama/Meta-Llama-3-8B-Instruct&quot;</span>)
prompt = <span class="hljs-string">&quot;Can I help&quot;</span>
inputs = tokenizer(prompt, return_tensors=<span class="hljs-string">&quot;pt&quot;</span>).input_ids.to(device)

<span class="hljs-comment"># distributed run</span>
outputs = model(inputs)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-968uev">Launch the inference script above on <a href="https://pytorch.org/docs/stable/elastic/run.html" rel="nofollow">torchrun</a> with 4 processes per GPU.</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-node 4 demo.py<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-8zwnyi">You can benefit from considerable speed ups for inference, especially for inputs with large batch size or long sequences.</p> <p data-svelte-h="svelte-bszveo">For a single forward pass on <a href="./model_doc/llama">Llama</a> with a sequence length of 512 and various batch sizes, you can expect the following speed ups.</p> <div style="text-align: center" data-svelte-h="svelte-lg3kfi"><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/Meta-Llama-3-8B-Instruct%2C%20seqlen%20%3D%20512%2C%20python%2C%20w_%20compile.png"></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/perf_infer_gpu_multi.md" target="_blank"><span data-svelte-h="svelte-1kd6by1">&lt;</span> <span data-svelte-h="svelte-x0xyl0">&gt;</span> <span data-svelte-h="svelte-1dajgef"><span class="underline ml-1.5">Update</span> on GitHub</span></a>  <p></p> 
			
			<script>
				{
					__sveltekit_1bm5psi = {
						assets: "/docs/transformers/pr_36839/en",
						base: "/docs/transformers/pr_36839/en",
						env: {}
					};

					const element = document.currentScript.parentElement;

					const data = [null,null];

					Promise.all([
						import("/docs/transformers/pr_36839/en/_app/immutable/entry/start.6be8d590.js"),
						import("/docs/transformers/pr_36839/en/_app/immutable/entry/app.09748b4b.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 415],
							data,
							form: null,
							error: null
						});
					});
				}
			</script>