transformers/pr_35010/en/quantization/finegrained_fp8.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Fine-grained FP8&quot;,&quot;local&quot;:&quot;fine-grained-fp8&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}">
		<link href="/docs/transformers/pr_35010/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/entry/start.80f0bf24.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/scheduler.01eeda35.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/singletons.2f9d3ffa.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/index.4862150a.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/paths.5ba5a4a8.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/entry/app.746bb5e5.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/index.6dd51b66.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/nodes/0.507a43fb.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/each.e59479a4.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/nodes/437.945aacd5.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/Tip.de9bae2b.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/CodeBlock.864da1b0.js">
		<link rel="modulepreload" href="/docs/transformers/pr_35010/en/_app/immutable/chunks/EditOnGithub.7faefd25.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Fine-grained FP8&quot;,&quot;local&quot;:&quot;fine-grained-fp8&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END -->      <p></p>   <h1 class="relative group"><a id="fine-grained-fp8" 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="#fine-grained-fp8"><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>Fine-grained FP8</span></h1> <p data-svelte-h="svelte-hrrhz7">Fine-grained FP8 quantization quantizes the weights and activations to fp8.</p> <ul data-svelte-h="svelte-1rx58b3"><li>The weights are quantized to 8-bits for each 2D block (<code>weight_block_size=(128, 128)</code>).</li> <li>The activations are quantized to 8-bits for each group per token. The group value matches the weights in the input channel (128 by default).</li></ul> <p data-svelte-h="svelte-10oa2n0">FP8 quantization enables support for <a href="https://hf.co/papers/2412.19437" rel="nofollow">DeepSeek-V3</a> and DeepSeek-R1.</p> <div class="flex justify-center" data-svelte-h="svelte-cd82ay"><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/b7b3b34bf826a6423ea82ffc57ecac80c46c3c76/transformers/quantization/quantization_deepseek.png"></div>  <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-aidjqb">You need a GPU with Compute Capability&gt;=9 (H100), and install a PyTorch version compatible with the CUDA version of your GPU.</p></div> <p data-svelte-h="svelte-1rty8u8">Install Accelerate and upgrade to the latest version of PyTorch.</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 -->pip install --upgrade accelerate torch<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-38h8ls">Create a <a href="/docs/transformers/pr_35010/en/main_classes/quantization#transformers.FineGrainedFP8Config">FineGrainedFP8Config</a> class and pass it to <a href="/docs/transformers/pr_35010/en/main_classes/model#transformers.PreTrainedModel.from_pretrained">from_pretrained()</a> to quantize it. The weights are loaded in full precision (<code>torch.float32</code>) by default regardless of the actual data type the weights are stored in. Set <code>torch_dtype=&quot;auto&quot;</code> to load the weights in the data type defined in a models <code>config.json</code> file to automatically load the most memory-optiomal data type.</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">from</span> transformers <span class="hljs-keyword">import</span> FineGrainedFP8Config, AutoModelForCausalLM, AutoTokenizer

model_name = <span class="hljs-string">&quot;meta-llama/Meta-Llama-3-8B&quot;</span>
quantization_config = FineGrainedFP8Config()
quantized_model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=<span class="hljs-string">&quot;auto&quot;</span>, device_map=<span class="hljs-string">&quot;auto&quot;</span>, quantization_config=quantization_config)

tokenizer = AutoTokenizer.from_pretrained(model_name)
input_text = <span class="hljs-string">&quot;What are we having for dinner?&quot;</span>
input_ids = tokenizer(input_text, return_tensors=<span class="hljs-string">&quot;pt&quot;</span>).to(<span class="hljs-string">&quot;cuda&quot;</span>)

output = quantized_model.generate(**input_ids, max_new_tokens=<span class="hljs-number">10</span>)
<span class="hljs-built_in">print</span>(tokenizer.decode(output[<span class="hljs-number">0</span>], skip_special_tokens=<span class="hljs-literal">True</span>))<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-wjzsdt">Use <a href="/docs/transformers/pr_35010/en/main_classes/model#transformers.PreTrainedModel.save_pretrained">save_pretrained()</a> to save the quantized model and reload it with <a href="/docs/transformers/pr_35010/en/main_classes/model#transformers.PreTrainedModel.from_pretrained">from_pretrained()</a>.</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5   text-gray-600 " title="code excerpt" type="button"><svg 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 -->quant_path = <span class="hljs-string">&quot;/path/to/save/quantized/model&quot;</span>
model.save_pretrained(quant_path)
model = AutoModelForCausalLM.from_pretrained(quant_path, device_map=<span class="hljs-string">&quot;auto&quot;</span>)<!-- 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/quantization/finegrained_fp8.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_14c1y9j = {
						assets: "/docs/transformers/pr_35010/en",
						base: "/docs/transformers/pr_35010/en",
						env: {}
					};

					const element = document.currentScript.parentElement;

					const data = [null,null];

					Promise.all([
						import("/docs/transformers/pr_35010/en/_app/immutable/entry/start.80f0bf24.js"),
						import("/docs/transformers/pr_35010/en/_app/immutable/entry/app.746bb5e5.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 437],
							data,
							form: null,
							error: null
						});
					});
				}
			</script>