diffusers/v0.3.0/en/optimization/fp16.html

<meta charset="utf-8" /><meta http-equiv="content-security-policy" content=""><meta name="hf:doc:metadata" content="{&quot;local&quot;:&quot;memory-and-speed&quot;,&quot;sections&quot;:[{&quot;local&quot;:&quot;cuda-autocast&quot;,&quot;title&quot;:&quot;CUDA `autocast`&quot;},{&quot;local&quot;:&quot;half-precision-weights&quot;,&quot;title&quot;:&quot;Half precision weights&quot;},{&quot;local&quot;:&quot;sliced-attention-for-additional-memory-savings&quot;,&quot;title&quot;:&quot;Sliced attention for additional memory savings&quot;}],&quot;title&quot;:&quot;Memory and speed&quot;}" data-svelte="svelte-1phssyn">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/assets/pages/__layout.svelte-hf-doc-builder.css">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/start-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/chunks/vendor-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/chunks/paths-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/pages/__layout.svelte-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/pages/optimization/fp16.mdx-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/chunks/Tip-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/chunks/IconCopyLink-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.3.0/en/_app/chunks/CodeBlock-hf-doc-builder.js"> 






<h1 class="relative group"><a id="memory-and-speed" 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="#memory-and-speed"><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>Memory and speed
	</span></h1>

<p>We present some techniques and ideas to optimize 🤗 Diffusers <em>inference</em> for memory or speed.</p>
<h2 class="relative group"><a id="cuda-autocast" 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="#cuda-autocast"><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>CUDA <code>autocast</code></span></h2>

<p>If you use a CUDA GPU, you can take advantage of <code>torch.autocast</code> to perform inference roughly twice as fast at the cost of slightly lower precision. All you need to do is put your inference call inside an <code>autocast</code> context manager. The following example shows how to do it using Stable Diffusion text-to-image generation as an example:</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><!-- HTML_TAG_START --><span class="hljs-keyword">from</span> torch <span class="hljs-keyword">import</span> autocast
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionPipeline

pipe = StableDiffusionPipeline.from_pretrained(<span class="hljs-string">&quot;CompVis/stable-diffusion-v1-4&quot;</span>, use_auth_token=<span class="hljs-literal">True</span>)
pipe = pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)

prompt = <span class="hljs-string">&quot;a photo of an astronaut riding a horse on mars&quot;</span>
<span class="hljs-keyword">with</span> autocast(<span class="hljs-string">&quot;cuda&quot;</span>):
    image = pipe(prompt).images[<span class="hljs-number">0</span>]  <!-- HTML_TAG_END --></pre></div>
<p>Despite the precision loss, in our experience the final image results look the same as the <code>float32</code> versions. Feel free to experiment and report back!</p>
<h2 class="relative group"><a id="half-precision-weights" 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="#half-precision-weights"><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>Half precision weights
	</span></h2>

<p>To save more GPU memory, you can load the model weights directly in half precision. This involves loading the float16 version of the weights, which was saved to a branch named <code>fp16</code>, and telling PyTorch to use the <code>float16</code> type when loading them:</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><!-- HTML_TAG_START -->pipe = StableDiffusionPipeline.from_pretrained(
    <span class="hljs-string">&quot;CompVis/stable-diffusion-v1-4&quot;</span>,
    revision=<span class="hljs-string">&quot;fp16&quot;</span>,
    torch_dtype=torch.float16,
    use_auth_token=<span class="hljs-literal">True</span>
)<!-- HTML_TAG_END --></pre></div>
<h2 class="relative group"><a id="sliced-attention-for-additional-memory-savings" 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="#sliced-attention-for-additional-memory-savings"><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>Sliced attention for additional memory savings
	</span></h2>

<p>For even additional memory savings, you can use a sliced version of attention that performs the computation in steps instead of all at once. </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">Attention slicing is useful even if a batch size of just 1 is used - as long as the model uses more than one attention head. If there is more than one attention head the *QK^T* attention matrix can be computed sequentially for each head which can save a significant amount of memory.
</div>
<p>To perform the attention computation sequentially over each head, you only need to invoke <a href="/docs/diffusers/v0.3.0/en/api/pipelines/stable_diffusion#diffusers.StableDiffusionPipeline.enable_attention_slicing">enable_attention_slicing()</a> in your pipeline before inference, like here:</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><!-- HTML_TAG_START --><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionPipeline

pipe = StableDiffusionPipeline.from_pretrained(
    <span class="hljs-string">&quot;CompVis/stable-diffusion-v1-4&quot;</span>,
    revision=<span class="hljs-string">&quot;fp16&quot;</span>,
    torch_dtype=torch.float16,
    use_auth_token=<span class="hljs-literal">True</span>
)
pipe = pipe.to(<span class="hljs-string">&quot;cuda&quot;</span>)

prompt = <span class="hljs-string">&quot;a photo of an astronaut riding a horse on mars&quot;</span>
pipe.enable_attention_slicing()
<span class="hljs-keyword">with</span> torch.autocast(<span class="hljs-string">&quot;cuda&quot;</span>):
    image = pipe(prompt).images[<span class="hljs-number">0</span>]  <!-- HTML_TAG_END --></pre></div>
<p>There’s a small performance penalty of about 10% slower inference times, but this method allows you to use Stable Diffusion in as little as 3.2 GB of VRAM! </p>


		<script type="module" data-hydrate="nd8sai">
		import { start } from "/docs/diffusers/v0.3.0/en/_app/start-hf-doc-builder.js";
		start({
			target: document.querySelector('[data-hydrate="nd8sai"]').parentNode,
			paths: {"base":"/docs/diffusers/v0.3.0/en","assets":"/docs/diffusers/v0.3.0/en"},
			session: {},
			route: false,
			spa: false,
			trailing_slash: "never",
			hydrate: {
				status: 200,
				error: null,
				nodes: [
					import("/docs/diffusers/v0.3.0/en/_app/pages/__layout.svelte-hf-doc-builder.js"),
						import("/docs/diffusers/v0.3.0/en/_app/pages/optimization/fp16.mdx-hf-doc-builder.js")
				],
				params: {}
			}
		});
	</script>