mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-22 03:35:00 +08:00
1447 lines
74 KiB
HTML
1447 lines
74 KiB
HTML
|
||
|
||
<!DOCTYPE html>
|
||
|
||
|
||
<html lang="en" data-content_root="../" >
|
||
|
||
<head>
|
||
<meta charset="utf-8" />
|
||
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
|
||
|
||
<title>Overview — TensorRT-LLM</title>
|
||
|
||
|
||
|
||
<script data-cfasync="false">
|
||
document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
|
||
document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
|
||
</script>
|
||
<!--
|
||
this give us a css class that will be invisible only if js is disabled
|
||
-->
|
||
<noscript>
|
||
<style>
|
||
.pst-js-only { display: none !important; }
|
||
|
||
</style>
|
||
</noscript>
|
||
|
||
<!-- Loaded before other Sphinx assets -->
|
||
<link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
|
||
<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
|
||
|
||
<link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
|
||
<link rel="stylesheet" type="text/css" href="../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
|
||
<link rel="stylesheet" type="text/css" href="../_static/copybutton.css?v=76b2166b" />
|
||
<link rel="stylesheet" type="text/css" href="../_static/autodoc_pydantic.css" />
|
||
<link rel="stylesheet" type="text/css" href="../_static/togglebutton.css?v=13237357" />
|
||
<link rel="stylesheet" type="text/css" href="../_static/custom.css?v=95073da6" />
|
||
|
||
<!-- So that users can add custom icons -->
|
||
<script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
|
||
<!-- Pre-loaded scripts that we'll load fully later -->
|
||
<link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
|
||
<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
|
||
|
||
<script src="../_static/documentation_options.js?v=5929fcd5"></script>
|
||
<script src="../_static/doctools.js?v=9a2dae69"></script>
|
||
<script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
|
||
<script src="../_static/clipboard.min.js?v=a7894cd8"></script>
|
||
<script src="../_static/copybutton.js?v=65e89d2a"></script>
|
||
<script>let toggleHintShow = 'Click to show';</script>
|
||
<script>let toggleHintHide = 'Click to hide';</script>
|
||
<script>let toggleOpenOnPrint = 'true';</script>
|
||
<script src="../_static/togglebutton.js?v=4a39c7ea"></script>
|
||
<script>var togglebuttonSelector = '.toggle, .admonition.dropdown';</script>
|
||
<script>var togglebuttonSelector = '.toggle, .admonition.dropdown';</script>
|
||
<script>DOCUMENTATION_OPTIONS.pagename = 'performance/perf-overview';</script>
|
||
<script>
|
||
DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
|
||
DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
|
||
DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.1.0rc5';
|
||
DOCUMENTATION_OPTIONS.show_version_warning_banner =
|
||
false;
|
||
</script>
|
||
<link rel="icon" href="../_static/favicon.png"/>
|
||
<link rel="index" title="Index" href="../genindex.html" />
|
||
<link rel="search" title="Search" href="../search.html" />
|
||
|
||
<meta name="viewport" content="width=device-width, initial-scale=1"/>
|
||
<meta name="docsearch:language" content="en"/>
|
||
<meta name="docsearch:version" content="1.1.0rc5" />
|
||
|
||
|
||
</head>
|
||
|
||
|
||
<body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
|
||
|
||
|
||
|
||
<div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
|
||
|
||
<div id="pst-scroll-pixel-helper"></div>
|
||
|
||
<button type="button" class="btn rounded-pill" id="pst-back-to-top">
|
||
<i class="fa-solid fa-arrow-up"></i>Back to top</button>
|
||
|
||
|
||
<dialog id="pst-search-dialog">
|
||
|
||
<form class="bd-search d-flex align-items-center"
|
||
action="../search.html"
|
||
method="get">
|
||
<i class="fa-solid fa-magnifying-glass"></i>
|
||
<input type="search"
|
||
class="form-control"
|
||
name="q"
|
||
placeholder="Search the docs ..."
|
||
aria-label="Search the docs ..."
|
||
autocomplete="off"
|
||
autocorrect="off"
|
||
autocapitalize="off"
|
||
spellcheck="false"/>
|
||
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
|
||
</form>
|
||
</dialog>
|
||
|
||
<div class="pst-async-banner-revealer d-none">
|
||
<aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
|
||
</div>
|
||
|
||
|
||
<header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
|
||
<div class="bd-header__inner bd-page-width">
|
||
<button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
|
||
<span class="fa-solid fa-bars"></span>
|
||
</button>
|
||
|
||
|
||
<div class="col-lg-3 navbar-header-items__start">
|
||
|
||
<div class="navbar-item">
|
||
|
||
|
||
|
||
|
||
|
||
<a class="navbar-brand logo" href="../index.html">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<img src="../_static/nvidia-logo-horiz-rgb-blk-for-screen.svg" class="logo__image only-light" alt="TensorRT-LLM - Home"/>
|
||
<img src="../_static/nvidia-logo-horiz-rgb-wht-for-screen.svg" class="logo__image only-dark pst-js-only" alt="TensorRT-LLM - Home"/>
|
||
|
||
|
||
<p class="title logo__title">TensorRT-LLM</p>
|
||
|
||
</a></div>
|
||
|
||
</div>
|
||
|
||
<div class="col-lg-9 navbar-header-items">
|
||
|
||
<div class="me-auto navbar-header-items__center">
|
||
|
||
<div class="navbar-item">
|
||
|
||
|
||
<div class="version-switcher__container dropdown pst-js-only">
|
||
<button id="pst-version-switcher-button-2"
|
||
type="button"
|
||
class="version-switcher__button btn btn-sm dropdown-toggle"
|
||
data-bs-toggle="dropdown"
|
||
aria-haspopup="listbox"
|
||
aria-controls="pst-version-switcher-list-2"
|
||
aria-label="Version switcher list"
|
||
>
|
||
Choose version <!-- this text may get changed later by javascript -->
|
||
<span class="caret"></span>
|
||
</button>
|
||
<div id="pst-version-switcher-list-2"
|
||
class="version-switcher__menu dropdown-menu list-group-flush py-0"
|
||
role="listbox" aria-labelledby="pst-version-switcher-button-2">
|
||
<!-- dropdown will be populated by javascript on page load -->
|
||
</div>
|
||
</div></div>
|
||
|
||
</div>
|
||
|
||
|
||
<div class="navbar-header-items__end">
|
||
|
||
<div class="navbar-item navbar-persistent--container">
|
||
|
||
|
||
<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
|
||
<i class="fa-solid fa-magnifying-glass"></i>
|
||
<span class="search-button__default-text">Search</span>
|
||
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
|
||
</button>
|
||
</div>
|
||
|
||
|
||
<div class="navbar-item">
|
||
|
||
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode" data-bs-placement="bottom" data-bs-toggle="tooltip">
|
||
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light" title="Light"></i>
|
||
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark" title="Dark"></i>
|
||
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto" title="System Settings"></i>
|
||
</button></div>
|
||
|
||
</div>
|
||
|
||
</div>
|
||
|
||
|
||
<div class="navbar-persistent--mobile">
|
||
|
||
<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
|
||
<i class="fa-solid fa-magnifying-glass"></i>
|
||
<span class="search-button__default-text">Search</span>
|
||
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
|
||
</button>
|
||
</div>
|
||
|
||
|
||
|
||
<button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
|
||
<span class="fa-solid fa-outdent"></span>
|
||
</button>
|
||
|
||
</div>
|
||
|
||
</header>
|
||
|
||
|
||
<div class="bd-container">
|
||
<div class="bd-container__inner bd-page-width">
|
||
|
||
|
||
|
||
<dialog id="pst-primary-sidebar-modal"></dialog>
|
||
<div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<a class="navbar-brand logo" href="../index.html">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<img src="../_static/nvidia-logo-horiz-rgb-blk-for-screen.svg" class="logo__image only-light" alt="TensorRT-LLM - Home"/>
|
||
<img src="../_static/nvidia-logo-horiz-rgb-wht-for-screen.svg" class="logo__image only-dark pst-js-only" alt="TensorRT-LLM - Home"/>
|
||
|
||
|
||
<p class="title logo__title">TensorRT-LLM</p>
|
||
|
||
</a>
|
||
|
||
|
||
|
||
<div class="sidebar-header-items sidebar-primary__section">
|
||
|
||
|
||
<div class="sidebar-header-items__center">
|
||
|
||
|
||
|
||
<div class="navbar-item">
|
||
|
||
|
||
<div class="version-switcher__container dropdown pst-js-only">
|
||
<button id="pst-version-switcher-button-3"
|
||
type="button"
|
||
class="version-switcher__button btn btn-sm dropdown-toggle"
|
||
data-bs-toggle="dropdown"
|
||
aria-haspopup="listbox"
|
||
aria-controls="pst-version-switcher-list-3"
|
||
aria-label="Version switcher list"
|
||
>
|
||
Choose version <!-- this text may get changed later by javascript -->
|
||
<span class="caret"></span>
|
||
</button>
|
||
<div id="pst-version-switcher-list-3"
|
||
class="version-switcher__menu dropdown-menu list-group-flush py-0"
|
||
role="listbox" aria-labelledby="pst-version-switcher-button-3">
|
||
<!-- dropdown will be populated by javascript on page load -->
|
||
</div>
|
||
</div></div>
|
||
|
||
|
||
</div>
|
||
|
||
|
||
|
||
<div class="sidebar-header-items__end">
|
||
|
||
<div class="navbar-item">
|
||
|
||
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode" data-bs-placement="bottom" data-bs-toggle="tooltip">
|
||
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light" title="Light"></i>
|
||
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark" title="Dark"></i>
|
||
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto" title="System Settings"></i>
|
||
</button></div>
|
||
|
||
</div>
|
||
|
||
</div>
|
||
|
||
<div class="sidebar-primary-items__start sidebar-primary__section">
|
||
<div class="sidebar-primary-item">
|
||
|
||
|
||
|
||
<nav class="bd-docs-nav bd-links"
|
||
aria-label="Table of Contents">
|
||
<p class="bd-links__title" role="heading" aria-level="1">Table of Contents</p>
|
||
<div class="bd-toc-item navbar-nav"><p aria-level="2" class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../overview.html">Overview</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../quick-start-guide.html">Quick Start Guide</a></li>
|
||
<li class="toctree-l1 has-children"><a class="reference internal" href="../installation/index.html">Installation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
|
||
<li class="toctree-l2"><a class="reference internal" href="../installation/containers.html">Pre-built release container images on NGC</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../installation/linux.html">Installing on Linux via <code class="docutils literal notranslate"><span class="pre">pip</span></code></a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../installation/build-from-source-linux.html">Building from Source Code on Linux</a></li>
|
||
</ul>
|
||
</details></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Deployment Guide</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1 has-children"><a class="reference internal" href="../examples/llm_api_examples.html">LLM Examples</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_inference.html">Generate text</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_inference_async.html">Generate text asynchronously</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_inference_async_streaming.html">Generate text in streaming</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_inference_distributed.html">Distributed LLM Generation</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_guided_decoding.html">Generate text with guided decoding</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_logits_processor.html">Control generated text using logits processor</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_multilora.html">Generate text with multiple LoRA adapters</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_speculative_decoding.html">Speculative Decoding</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_kv_cache_connector.html">KV Cache Connector</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_runtime.html">Runtime Configuration Examples</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_sampling.html">Sampling Techniques Showcase</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_mgmn_llm_distributed.html">Run LLM-API with pytorch backend on Slurm</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_mgmn_trtllm_bench.html">Run trtllm-bench with pytorch backend on Slurm</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/llm_mgmn_trtllm_serve.html">Run trtllm-serve with pytorch backend on Slurm</a></li>
|
||
</ul>
|
||
</details></li>
|
||
<li class="toctree-l1 has-children"><a class="reference internal" href="../examples/trtllm_serve_examples.html">Online Serving Examples</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/curl_chat_client.html">Curl Chat Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/curl_chat_client_for_multimodal.html">Curl Chat Client For Multimodal</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/curl_completion_client.html">Curl Completion Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/deepseek_r1_reasoning_parser.html">Deepseek R1 Reasoning Parser</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/genai_perf_client.html">Genai Perf Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/genai_perf_client_for_multimodal.html">Genai Perf Client For Multimodal</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/openai_chat_client.html">OpenAI Chat Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/openai_chat_client_for_multimodal.html">OpenAI Chat Client for Multimodal</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/openai_completion_client.html">OpenAI Completion Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/openai_completion_client_for_lora.html">Openai Completion Client For Lora</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../examples/openai_completion_client_json_schema.html">OpenAI Completion Client with JSON Schema</a></li>
|
||
</ul>
|
||
</details></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../examples/dynamo_k8s_example.html">Dynamo K8s Example</a></li>
|
||
<li class="toctree-l1 has-children"><a class="reference internal" href="../deployment-guide/index.html">Model Recipes</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
|
||
<li class="toctree-l2"><a class="reference internal" href="../deployment-guide/quick-start-recipe-for-deepseek-r1-on-trtllm.html">Quick Start Recipe for DeepSeek R1 on TensorRT LLM - Blackwell & Hopper Hardware</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../deployment-guide/quick-start-recipe-for-llama3.3-70b-on-trtllm.html">Quick Start Recipe for Llama3.3 70B on TensorRT LLM - Blackwell & Hopper Hardware</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../deployment-guide/quick-start-recipe-for-llama4-scout-on-trtllm.html">Quick Start Recipe for Llama4 Scout 17B on TensorRT LLM - Blackwell & Hopper Hardware</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../deployment-guide/quick-start-recipe-for-gpt-oss-on-trtllm.html">Quick Start Recipe for GPT-OSS on TensorRT-LLM - Blackwell Hardware</a></li>
|
||
</ul>
|
||
</details></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Models</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../models/supported-models.html">Supported Models</a></li>
|
||
|
||
<li class="toctree-l1"><a class="reference internal" href="../models/adding-new-model.html">Adding a New Model</a></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">CLI Reference</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../commands/trtllm-bench.html">trtllm-bench</a></li>
|
||
|
||
<li class="toctree-l1"><a class="reference internal" href="../commands/trtllm-eval.html">trtllm-eval</a></li>
|
||
<li class="toctree-l1 has-children"><a class="reference internal" href="../commands/trtllm-serve/index.html">trtllm-serve</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
|
||
<li class="toctree-l2"><a class="reference internal" href="../commands/trtllm-serve/trtllm-serve.html">trtllm-serve</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="../commands/trtllm-serve/run-benchmark-with-trtllm-serve.html">Run benchmarking with <code class="docutils literal notranslate"><span class="pre">trtllm-serve</span></code></a></li>
|
||
</ul>
|
||
</details></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">API Reference</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../llm-api/index.html">LLM API Introduction</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../llm-api/reference.html">API Reference</a></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Features</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/feature-combination-matrix.html">Feature Combination Matrix</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/attention.html">Multi-Head, Multi-Query, and Group-Query Attention</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/disagg-serving.html">Disaggregated Serving (Beta)</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/kvcache.html">KV Cache System</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/long-sequence.html">Long Sequences</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/lora.html">LoRA (Low-Rank Adaptation)</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/multi-modality.html">Multimodal Support in TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/overlap-scheduler.html">Overlap Scheduler</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/paged-attention-ifb-scheduler.html">Paged Attention, IFB, and Request Scheduling</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/parallel-strategy.html">Parallelism in TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/quantization.html">Quantization</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/sampling.html">Sampling</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/speculative-decoding.html">Speculative Decoding</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/checkpoint-loading.html">Checkpoint Loading</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../features/auto_deploy/auto-deploy.html">AutoDeploy (Prototype)</a></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Developer Guide</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../architecture/overview.html">Architecture Overview</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../developer-guide/perf-analysis.html">Performance Analysis</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../developer-guide/perf-benchmarking.html">TensorRT LLM Benchmarking</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../developer-guide/ci-overview.html">Continuous Integration Overview</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../developer-guide/dev-containers.html">Using Dev Containers</a></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Blogs</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog10_ADP_Balance_Strategy.html">ADP Balance Strategy</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog11_GPT_OSS_Eagle3.html">Running GPT-OSS-120B with Eagle3 Speculative Decoding on GB200/B200 (TensorRT LLM)</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog1_Pushing_Latency_Boundaries_Optimizing_DeepSeek-R1_Performance_on_NVIDIA_B200_GPUs.html">Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.html">DeepSeek R1 MTP Implementation and Optimization</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs.html">Optimizing DeepSeek R1 Throughput on NVIDIA Blackwell GPUs: A Deep Dive for Developers</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog4_Scaling_Expert_Parallelism_in_TensorRT-LLM.html">Scaling Expert Parallelism in TensorRT LLM (Part 1: Design and Implementation of Large-scale EP)</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog5_Disaggregated_Serving_in_TensorRT-LLM.html">Disaggregated Serving in TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog6_Llama4_maverick_eagle_guide.html">How to launch Llama4 Maverick + Eagle3 TensorRT LLM server</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog7_NGram_performance_Analysis_And_Auto_Enablement.html">N-Gram Speculative Decoding in TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog8_Scaling_Expert_Parallelism_in_TensorRT-LLM_part2.html">Scaling Expert Parallelism in TensorRT LLM (Part 2: Performance Status and Optimization)</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/tech_blog/blog9_Deploying_GPT_OSS_on_TRTLLM.html">Running a High Performance GPT-OSS-120B Inference Server with TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.html">How to get best performance on DeepSeek-R1 in TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/H200launch.html">H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT LLM</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/XQA-kernel.html">New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../blogs/H100vsA100.html">H100 has 4.6x A100 Performance in TensorRT LLM, achieving 10,000 tok/s at 100ms to first token</a></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Quick Links</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference external" href="https://github.com/NVIDIA/TensorRT-LLM/releases">Releases</a></li>
|
||
<li class="toctree-l1"><a class="reference external" href="https://github.com/NVIDIA/TensorRT-LLM">Github Code</a></li>
|
||
<li class="toctree-l1"><a class="reference external" href="https://github.com/NVIDIA/TensorRT-LLM/issues?q=is%3Aissue%20state%3Aopen%20label%3Aroadmap">Roadmap</a></li>
|
||
</ul>
|
||
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Use TensorRT Engine</span></p>
|
||
<ul class="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../legacy/tensorrt_quickstart.html">LLM API with TensorRT Engine</a></li>
|
||
</ul>
|
||
</div>
|
||
</nav></div>
|
||
</div>
|
||
|
||
|
||
<div class="sidebar-primary-items__end sidebar-primary__section">
|
||
</div>
|
||
|
||
|
||
|
||
</div>
|
||
|
||
<main id="main-content" class="bd-main" role="main">
|
||
|
||
|
||
<div class="bd-content">
|
||
<div class="bd-article-container">
|
||
|
||
<div class="bd-header-article d-print-none">
|
||
<div class="header-article-items header-article__inner">
|
||
|
||
<div class="header-article-items__start">
|
||
|
||
<div class="header-article-item">
|
||
|
||
<nav aria-label="Breadcrumb" class="d-print-none">
|
||
<ul class="bd-breadcrumbs">
|
||
|
||
<li class="breadcrumb-item breadcrumb-home">
|
||
<a href="../index.html" class="nav-link" aria-label="Home">
|
||
<i class="fa-solid fa-home"></i>
|
||
</a>
|
||
</li>
|
||
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Overview</span></li>
|
||
</ul>
|
||
</nav>
|
||
</div>
|
||
|
||
</div>
|
||
|
||
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
<div id="searchbox"></div>
|
||
<article class="bd-article">
|
||
|
||
<section id="overview">
|
||
<span id="perf-overview"></span><h1>Overview<a class="headerlink" href="#overview" title="Link to this heading">#</a></h1>
|
||
<p>This document summarizes performance measurements of TensorRT-LLM on a number of GPUs across a set of key models.</p>
|
||
<p>The data in the following tables is provided as a reference point to help users validate observed performance.
|
||
It should <em>not</em> be considered as the peak performance that can be delivered by TensorRT-LLM.</p>
|
||
<p>We attempted to keep commands as simple as possible to ease reproducibility and left many options at their default settings.
|
||
Tuning batch sizes, parallelism configurations, and other options may lead to improved performance depending on your situaiton.</p>
|
||
<p>For DeepSeek R1 performance, please check out our <a class="reference internal" href="../blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.html"><span class="std std-doc">performance guide</span></a></p>
|
||
<p>For more information on benchmarking with <code class="docutils literal notranslate"><span class="pre">trtllm-bench</span></code> see this NVIDIA <a class="reference external" href="https://developer.nvidia.com/blog/llm-inference-benchmarking-performance-tuning-with-tensorrt-llm/">blog post</a>.</p>
|
||
<section id="throughput-measurements">
|
||
<h2>Throughput Measurements<a class="headerlink" href="#throughput-measurements" title="Link to this heading">#</a></h2>
|
||
<p>The below table shows performance data where a local inference client is fed requests at an infinite rate (no delay between messages),
|
||
and shows the throughput scenario under maximum load. The reported metric is <code class="docutils literal notranslate"><span class="pre">Total</span> <span class="pre">Output</span> <span class="pre">Throughput</span> <span class="pre">(tokens/sec)</span></code>.</p>
|
||
<p>The performance numbers below were collected using the steps described in this document.</p>
|
||
<p>Testing was performed on models with weights quantized using <a class="reference external" href="https://nvidia.github.io/TensorRT-Model-Optimizer/#">ModelOpt</a> and published by NVIDIA on the <a class="reference external" href="https://huggingface.co/collections/nvidia/model-optimizer-66aa84f7966b3150262481a4">Model Optimizer HuggingFace Collection</a>.</p>
|
||
<section id="hardware">
|
||
<h3>Hardware<a class="headerlink" href="#hardware" title="Link to this heading">#</a></h3>
|
||
<p>The following GPU variants were used for testing:</p>
|
||
<ul class="simple">
|
||
<li><p>H100 SXM 80GB (DGX H100)</p></li>
|
||
<li><p>H200 SXM 141GB (DGX H200)</p></li>
|
||
<li><p>GH200 96GB HBM3 (480GB LPDDR5X)</p></li>
|
||
<li><p>B200 180GB (DGX B200)</p></li>
|
||
<li><p>GB200 192GB (GB200 NVL72)</p></li>
|
||
</ul>
|
||
<p>Other hardware variants may have different TDP, memory bandwidth, core count, or other features leading to performance differences on these workloads.</p>
|
||
</section>
|
||
<section id="fp4-models">
|
||
<h3>FP4 Models<a class="headerlink" href="#fp4-models" title="Link to this heading">#</a></h3>
|
||
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>nvidia/Llama-3.3-70B-Instruct-FP4
|
||
nvidia/Llama-3.1-405B-Instruct-FP4
|
||
</pre></div>
|
||
</div>
|
||
<section id="llama-3-3-70b-fp4">
|
||
<h4>Llama 3.3 70B FP4<a class="headerlink" href="#llama-3-3-70b-fp4" title="Link to this heading">#</a></h4>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p></p></th>
|
||
<th class="head text-left"><p>GPU:</p></th>
|
||
<th class="head text-left"><p>B200</p></th>
|
||
<th class="head text-left"><p>GB200</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>TP Size</p></td>
|
||
<td class="text-left"><p>1</p></td>
|
||
<td class="text-left"><p>1</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>ISL, OSL</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>10,613.84</p></td>
|
||
<td class="text-left"><p>11,100.97</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>128, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>9,445.51</p></td>
|
||
<td class="text-left"><p>10,276.05</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 4096</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,276.85</p></td>
|
||
<td class="text-left"><p>7,351.12</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>500, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,983.27</p></td>
|
||
<td class="text-left"><p>8,194.30</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1000, 1000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,434.29</p></td>
|
||
<td class="text-left"><p>7,401.80</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>1000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,725.03</p></td>
|
||
<td class="text-left"><p>6,478.72</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1024, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,546.61</p></td>
|
||
<td class="text-left"><p>7,922.88</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>2048, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>1,330.35</p></td>
|
||
<td class="text-left"><p>1,418.47</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>2048, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,528.48</p></td>
|
||
<td class="text-left"><p>5,326.77</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>5000, 500</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>1,427.44</p></td>
|
||
<td class="text-left"><p>1,502.44</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>20000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>636.36</p></td>
|
||
<td class="text-left"><p>732.43</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="llama-3-1-405b-fp4">
|
||
<h4>Llama 3.1 405B FP4<a class="headerlink" href="#llama-3-1-405b-fp4" title="Link to this heading">#</a></h4>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p></p></th>
|
||
<th class="head text-left"><p>GPU:</p></th>
|
||
<th class="head text-left"><p>B200</p></th>
|
||
<th class="head text-left"><p>GB200</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>TP Size</p></td>
|
||
<td class="text-left"><p>4</p></td>
|
||
<td class="text-left"><p>4</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>ISL, OSL</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,218.89</p></td>
|
||
<td class="text-left"><p>6,598.97</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>128, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>7,178.10</p></td>
|
||
<td class="text-left"><p>7,497.40</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 4096</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>5,890.89</p></td>
|
||
<td class="text-left"><p>5,898.19</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>500, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>5,844.37</p></td>
|
||
<td class="text-left"><p>6,198.33</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1000, 1000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,958.53</p></td>
|
||
<td class="text-left"><p>5,243.35</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>1000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,874.16</p></td>
|
||
<td class="text-left"><p>4,905.51</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1024, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,833.19</p></td>
|
||
<td class="text-left"><p>4,686.38</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>2048, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>737.95</p></td>
|
||
<td class="text-left"><p>761.58</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>2048, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,024.02</p></td>
|
||
<td class="text-left"><p>4,326.56</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>5000, 500</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>1,032.40</p></td>
|
||
<td class="text-left"><p>1,078.87</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>20000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>667.39</p></td>
|
||
<td class="text-left"><p>649.95</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
</section>
|
||
<section id="fp8-models">
|
||
<h3>FP8 Models<a class="headerlink" href="#fp8-models" title="Link to this heading">#</a></h3>
|
||
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>nvidia/Llama-3.1-8B-Instruct-FP8
|
||
nvidia/Llama-3.3-70B-Instruct-FP8
|
||
nvidia/Llama-3.1-405B-Instruct-FP8
|
||
nvidia/Llama-4-Maverick-17B-128E-Instruct-FP8
|
||
</pre></div>
|
||
</div>
|
||
<section id="llama-3-1-8b-fp8">
|
||
<h4>Llama 3.1 8B FP8<a class="headerlink" href="#llama-3-1-8b-fp8" title="Link to this heading">#</a></h4>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p></p></th>
|
||
<th class="head text-left"><p>GPU:</p></th>
|
||
<th class="head text-left"><p>GH200</p></th>
|
||
<th class="head text-left"><p>H100</p></th>
|
||
<th class="head text-left"><p>H200</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>TP Size</p></td>
|
||
<td class="text-left"><p>1</p></td>
|
||
<td class="text-left"><p>1</p></td>
|
||
<td class="text-left"><p>1</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>ISL, OSL</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>27,304.25</p></td>
|
||
<td class="text-left"><p>26,401.48</p></td>
|
||
<td class="text-left"><p>27,027.80</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>128, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>24,045.60</p></td>
|
||
<td class="text-left"><p>21,413.21</p></td>
|
||
<td class="text-left"><p>23,102.25</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 4096</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>15,409.85</p></td>
|
||
<td class="text-left"><p>13,541.54</p></td>
|
||
<td class="text-left"><p>17,396.83</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>500, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>20,123.88</p></td>
|
||
<td class="text-left"><p>17,571.01</p></td>
|
||
<td class="text-left"><p>19,759.16</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1000, 1000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>16,352.99</p></td>
|
||
<td class="text-left"><p>14,991.62</p></td>
|
||
<td class="text-left"><p>17,162.49</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>1000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>15,705.82</p></td>
|
||
<td class="text-left"><p>13,505.23</p></td>
|
||
<td class="text-left"><p>16,227.11</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1024, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>16,102.52</p></td>
|
||
<td class="text-left"><p>13,165.91</p></td>
|
||
<td class="text-left"><p>16,057.66</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>2048, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,573.85</p></td>
|
||
<td class="text-left"><p>3,275.55</p></td>
|
||
<td class="text-left"><p>3,390.69</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>2048, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>10,767.05</p></td>
|
||
<td class="text-left"><p>9,462.43</p></td>
|
||
<td class="text-left"><p>11,822.14</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>5000, 500</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,584.74</p></td>
|
||
<td class="text-left"><p>3,276.47</p></td>
|
||
<td class="text-left"><p>3,758.08</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>20000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>1,393.31</p></td>
|
||
<td class="text-left"><p>1,340.69</p></td>
|
||
<td class="text-left"><p>1,705.68</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="llama-3-3-70b-fp8">
|
||
<h4>Llama 3.3 70B FP8<a class="headerlink" href="#llama-3-3-70b-fp8" title="Link to this heading">#</a></h4>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p></p></th>
|
||
<th class="head text-left"><p>GPU:</p></th>
|
||
<th class="head text-left"><p>H100</p></th>
|
||
<th class="head text-left"><p>H200</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>TP Size</p></td>
|
||
<td class="text-left"><p>2</p></td>
|
||
<td class="text-left"><p>2</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>ISL, OSL</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>6,092.28</p></td>
|
||
<td class="text-left"><p>6,327.98</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>128, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>5,892.94</p></td>
|
||
<td class="text-left"><p>7,467.36</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 4096</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,828.46</p></td>
|
||
<td class="text-left"><p>5,526.42</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>500, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,654.74</p></td>
|
||
<td class="text-left"><p>6,639.15</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1000, 1000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,181.06</p></td>
|
||
<td class="text-left"><p>4,773.33</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>1000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,708.93</p></td>
|
||
<td class="text-left"><p>5,790.36</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1024, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,785.04</p></td>
|
||
<td class="text-left"><p>5,480.44</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>2048, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>723.40</p></td>
|
||
<td class="text-left"><p>747.55</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>2048, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>2,785.53</p></td>
|
||
<td class="text-left"><p>3,775.80</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>5000, 500</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>865.55</p></td>
|
||
<td class="text-left"><p>978.28</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>20000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>411.85</p></td>
|
||
<td class="text-left"><p>609.42</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="llama-3-1-405b-fp8">
|
||
<h4>Llama 3.1 405B FP8<a class="headerlink" href="#llama-3-1-405b-fp8" title="Link to this heading">#</a></h4>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p></p></th>
|
||
<th class="head text-left"><p>GPU:</p></th>
|
||
<th class="head text-left"><p>H100</p></th>
|
||
<th class="head text-left"><p>H200</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>TP Size</p></td>
|
||
<td class="text-left"><p>8</p></td>
|
||
<td class="text-left"><p>8</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>Runtime Input/Output Lengths</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,705.18</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>128, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,517.39</p></td>
|
||
<td class="text-left"><p>4,715.13</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 4096</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>2,910.31</p></td>
|
||
<td class="text-left"><p>4,475.91</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>500, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,664.62</p></td>
|
||
<td class="text-left"><p>4,804.10</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1000, 1000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>2,955.50</p></td>
|
||
<td class="text-left"><p>3,208.25</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>1000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>2,884.69</p></td>
|
||
<td class="text-left"><p>3,630.29</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1024, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>3,237.41</p></td>
|
||
<td class="text-left"><p>3,609.50</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>2048, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>433.47</p></td>
|
||
<td class="text-left"><p>441.35</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>2048, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>2,216.55</p></td>
|
||
<td class="text-left"><p>2,840.86</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>5000, 500</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>579.05</p></td>
|
||
<td class="text-left"><p>645.26</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>20000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>363.27</p></td>
|
||
<td class="text-left"><p>509.87</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="llama-4-maverick-fp8">
|
||
<h4>Llama 4 Maverick FP8<a class="headerlink" href="#llama-4-maverick-fp8" title="Link to this heading">#</a></h4>
|
||
<p>Note: Performance for Llama 4 on sequence lengths less than 8,192 tokens is affected by an issue introduced in v0.21. To reproduce the Llama 4 performance noted here, please use v0.20</p>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p></p></th>
|
||
<th class="head text-left"><p>GPU</p></th>
|
||
<th class="head text-left"><p>H200</p></th>
|
||
<th class="head text-left"><p>H100</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>TP Size</p></td>
|
||
<td class="text-left"><p>8</p></td>
|
||
<td class="text-left"><p>8</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>ISL, OSL</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>128, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>27,543.87</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>128, 4096</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>18,541.01</p></td>
|
||
<td class="text-left"><p>11,163.12</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>500, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>21,117.34</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>1000, 2000</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>10,556.00</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>1024, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>16,859.45</p></td>
|
||
<td class="text-left"><p>11,584.33</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>2048, 128</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>4,364.06</p></td>
|
||
<td class="text-left"><p>3,832.38</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p>2048, 2048</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>12,800.89</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p>5000, 500</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
<td class="text-left"><p>5,128.60</p></td>
|
||
<td class="text-left"><p></p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
</section>
|
||
</section>
|
||
<section id="reproducing-benchmarked-results">
|
||
<h2>Reproducing Benchmarked Results<a class="headerlink" href="#reproducing-benchmarked-results" title="Link to this heading">#</a></h2>
|
||
<blockquote>
|
||
<div><p>[!NOTE] The only models supported in this workflow are those listed in the table above.</p>
|
||
</div></blockquote>
|
||
<p>The following tables are references for commands that are used as part of the benchmarking process. For a more detailed
|
||
description of this benchmarking workflow, see the <a class="reference external" href="https://nvidia.github.io/TensorRT-LLM/performance/perf-benchmarking.html">benchmarking suite documentation</a>.</p>
|
||
<section id="command-overview">
|
||
<h3>Command Overview<a class="headerlink" href="#command-overview" title="Link to this heading">#</a></h3>
|
||
<p>Starting with v0.19, testing was performed using the PyTorch backend - this workflow does not require an engine to be built.</p>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p>Stage</p></th>
|
||
<th class="head"><p>Description</p></th>
|
||
<th class="head"><p>Command</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p><a class="reference internal" href="#preparing-a-dataset">Dataset</a></p></td>
|
||
<td><p>Create a synthetic dataset</p></td>
|
||
<td><p><code class="docutils literal notranslate"><span class="pre">python</span> <span class="pre">benchmarks/cpp/prepare_dataset.py</span> <span class="pre">--tokenizer=$model_name</span> <span class="pre">--stdout</span> <span class="pre">token-norm-dist</span> <span class="pre">--num-requests=$num_requests</span> <span class="pre">--input-mean=$isl</span> <span class="pre">--output-mean=$osl</span> <span class="pre">--input-stdev=0</span> <span class="pre">--output-stdev=0</span> <span class="pre">></span> <span class="pre">$dataset_file</span></code></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p><a class="reference internal" href="#running-the-benchmark">Run</a></p></td>
|
||
<td><p>Run a benchmark with a dataset</p></td>
|
||
<td><p><code class="docutils literal notranslate"><span class="pre">trtllm-bench</span> <span class="pre">--model</span> <span class="pre">$model_name</span> <span class="pre">throughput</span> <span class="pre">--dataset</span> <span class="pre">$dataset_file</span> <span class="pre">--backend</span> <span class="pre">pytorch</span> <span class="pre">--extra_llm_api_options</span> <span class="pre">$llm_options</span></code></p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="variables">
|
||
<h3>Variables<a class="headerlink" href="#variables" title="Link to this heading">#</a></h3>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-left"><p>Name</p></th>
|
||
<th class="head"><p>Description</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$isl</span></code></p></td>
|
||
<td><p>Benchmark input sequence length.</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$osl</span></code></p></td>
|
||
<td><p>Benchmark output sequence length.</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$tp_size</span></code></p></td>
|
||
<td><p>Tensor parallel mapping degree to run the benchmark with</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$pp_size</span></code></p></td>
|
||
<td><p>Pipeline parallel mapping degree to run the benchmark with</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$model_name</span></code></p></td>
|
||
<td><p>HuggingFace model name eg. meta-llama/Llama-2-7b-hf or use the path to a local weights directory</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$dataset_file</span></code></p></td>
|
||
<td><p>Location of the dataset file generated by <code class="docutils literal notranslate"><span class="pre">prepare_dataset.py</span></code></p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$num_requests</span></code></p></td>
|
||
<td><p>The number of requests to generate for dataset generation</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$seq_len</span></code></p></td>
|
||
<td><p>A sequence length of ISL + OSL</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td class="text-left"><p><code class="docutils literal notranslate"><span class="pre">$llm_options</span></code></p></td>
|
||
<td><p>(optional) A yaml file containing additional options for the LLM API</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="preparing-a-dataset">
|
||
<h3>Preparing a Dataset<a class="headerlink" href="#preparing-a-dataset" title="Link to this heading">#</a></h3>
|
||
<p>In order to prepare a dataset, you can use the provided <a class="reference download internal" download="" href="../_downloads/ea8faa5e98124e92f96b66dc586fb429/prepare_dataset.py"><span class="xref download myst">script</span></a>.
|
||
To generate a synthetic dataset, run the following command:</p>
|
||
<div class="highlight-shell notranslate"><div class="highlight"><pre><span></span>python<span class="w"> </span>benchmarks/cpp/prepare_dataset.py<span class="w"> </span>--tokenizer<span class="o">=</span><span class="nv">$model_name</span><span class="w"> </span>--stdout<span class="w"> </span>token-norm-dist<span class="w"> </span>--num-requests<span class="o">=</span><span class="nv">$num_requests</span><span class="w"> </span>--input-mean<span class="o">=</span><span class="nv">$isl</span><span class="w"> </span>--output-mean<span class="o">=</span><span class="nv">$osl</span><span class="w"> </span>--input-stdev<span class="o">=</span><span class="m">0</span><span class="w"> </span>--output-stdev<span class="o">=</span><span class="m">0</span><span class="w"> </span>><span class="w"> </span><span class="nv">$dataset_file</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>The command will generate a text file located at the path specified <code class="docutils literal notranslate"><span class="pre">$dataset_file</span></code> where all requests are of the same
|
||
input/output sequence length combinations. The script works by using the tokenizer to retrieve the vocabulary size and
|
||
randomly sample token IDs from it to create entirely random sequences. In the command above, all requests will be uniform
|
||
because the standard deviations for both input and output sequences are set to 0.</p>
|
||
<p>For each input and output sequence length combination, the table below details the <code class="docutils literal notranslate"><span class="pre">$num_requests</span></code> that were used. For
|
||
shorter input and output lengths, a larger number of messages were used to guarantee that the system hit a steady state
|
||
because requests enter and exit the system at a much faster rate. For longer input/output sequence lengths, requests
|
||
remain in the system longer and therefore require less requests to achieve steady state.</p>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head"><p>Input Length</p></th>
|
||
<th class="head"><p>Output Length</p></th>
|
||
<th class="head"><p>$seq_len</p></th>
|
||
<th class="head"><p>$num_requests</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td><p>128</p></td>
|
||
<td><p>128</p></td>
|
||
<td><p>256</p></td>
|
||
<td><p>30000</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td><p>128</p></td>
|
||
<td><p>2048</p></td>
|
||
<td><p>2176</p></td>
|
||
<td><p>3000</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td><p>128</p></td>
|
||
<td><p>4096</p></td>
|
||
<td><p>4224</p></td>
|
||
<td><p>1500</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td><p>1000</p></td>
|
||
<td><p>2000</p></td>
|
||
<td><p>3000</p></td>
|
||
<td><p>1500</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td><p>2048</p></td>
|
||
<td><p>128</p></td>
|
||
<td><p>2176</p></td>
|
||
<td><p>3000</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td><p>2048</p></td>
|
||
<td><p>2048</p></td>
|
||
<td><p>4096</p></td>
|
||
<td><p>1500</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td><p>5000</p></td>
|
||
<td><p>500</p></td>
|
||
<td><p>5500</p></td>
|
||
<td><p>1500</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td><p>1000</p></td>
|
||
<td><p>1000</p></td>
|
||
<td><p>2000</p></td>
|
||
<td><p>3000</p></td>
|
||
</tr>
|
||
<tr class="row-even"><td><p>500</p></td>
|
||
<td><p>2000</p></td>
|
||
<td><p>2500</p></td>
|
||
<td><p>3000</p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td><p>20000</p></td>
|
||
<td><p>2000</p></td>
|
||
<td><p>22000</p></td>
|
||
<td><p>1000</p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</section>
|
||
<section id="running-the-benchmark">
|
||
<h3>Running the Benchmark<a class="headerlink" href="#running-the-benchmark" title="Link to this heading">#</a></h3>
|
||
<p>To run the benchmark with the generated data set, simply use the <code class="docutils literal notranslate"><span class="pre">trtllm-bench</span> <span class="pre">throughput</span></code> subcommand. The benchmarker will
|
||
run an offline maximum throughput scenario such that all requests are queued in rapid succession. You simply need to provide
|
||
a model name (HuggingFace reference or path to a local model), a <a class="reference internal" href="#preparing-a-dataset">generated dataset</a>, and a file containing any desired extra options to the LLMApi (details in <a class="reference download internal" download="" href="../_downloads/cba6509356738d5d6b4dcb3b7f52cf39/llm_args.py"><span class="xref download myst">tensorrt_llm/llmapi/llm_args.py:LlmArgs</span></a>).</p>
|
||
<div class="highlight-shell notranslate"><div class="highlight"><pre><span></span>trtllm-bench<span class="w"> </span>--model<span class="w"> </span><span class="nv">$model_name</span><span class="w"> </span>throughput<span class="w"> </span>--dataset<span class="w"> </span><span class="nv">$dataset_file</span><span class="w"> </span>--backend<span class="w"> </span>pytorch<span class="w"> </span>--extra_llm_api_options<span class="w"> </span><span class="nv">$llm_options</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>The data collected for the v0.21 benchmarks was run with the following file:</p>
|
||
<p><code class="docutils literal notranslate"><span class="pre">llm_options.yml</span></code></p>
|
||
<div class="highlight-yaml notranslate"><div class="highlight"><pre><span></span><span class="nt">cuda_graph_config</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="nt">enable_padding</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">true</span>
|
||
<span class="w"> </span><span class="nt">batch_sizes</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">1</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">2</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">4</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">8</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">16</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">32</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">64</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">128</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">256</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">384</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">512</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">1024</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">2048</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">4096</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">8192</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>In many cases, we also use a higher KV cache percentage by setting <code class="docutils literal notranslate"><span class="pre">--kv_cache_free_gpu_mem_fraction</span> <span class="pre">0.95</span></code> in the benchmark command. This allows us to obtain better performance than the default setting of <code class="docutils literal notranslate"><span class="pre">0.90</span></code>. We fall back to <code class="docutils literal notranslate"><span class="pre">0.90</span></code> or lower if out-of-memory errors are encountered.</p>
|
||
<p>The results will be printed to the terminal upon benchmark completion. For example,</p>
|
||
<div class="highlight-shell notranslate"><div class="highlight"><pre><span></span><span class="o">===========================================================</span>
|
||
<span class="o">=</span><span class="w"> </span>PERFORMANCE<span class="w"> </span><span class="nv">OVERVIEW</span>
|
||
<span class="o">===========================================================</span>
|
||
Request<span class="w"> </span>Throughput<span class="w"> </span><span class="o">(</span>req/sec<span class="o">)</span>:<span class="w"> </span><span class="m">43</span>.2089
|
||
Total<span class="w"> </span>Output<span class="w"> </span>Throughput<span class="w"> </span><span class="o">(</span>tokens/sec<span class="o">)</span>:<span class="w"> </span><span class="m">5530</span>.7382
|
||
Per<span class="w"> </span>User<span class="w"> </span>Output<span class="w"> </span>Throughput<span class="w"> </span><span class="o">(</span>tokens/sec/user<span class="o">)</span>:<span class="w"> </span><span class="m">2</span>.0563
|
||
Per<span class="w"> </span>GPU<span class="w"> </span>Output<span class="w"> </span>Throughput<span class="w"> </span><span class="o">(</span>tokens/sec/gpu<span class="o">)</span>:<span class="w"> </span><span class="m">5530</span>.7382
|
||
Total<span class="w"> </span>Token<span class="w"> </span>Throughput<span class="w"> </span><span class="o">(</span>tokens/sec<span class="o">)</span>:<span class="w"> </span><span class="m">94022</span>.5497
|
||
Total<span class="w"> </span>Latency<span class="w"> </span><span class="o">(</span>ms<span class="o">)</span>:<span class="w"> </span><span class="m">115716</span>.9214
|
||
Average<span class="w"> </span>request<span class="w"> </span>latency<span class="w"> </span><span class="o">(</span>ms<span class="o">)</span>:<span class="w"> </span><span class="m">75903</span>.4456
|
||
Per<span class="w"> </span>User<span class="w"> </span>Output<span class="w"> </span>Speed<span class="w"> </span><span class="o">[</span><span class="m">1</span>/TPOT<span class="o">]</span><span class="w"> </span><span class="o">(</span>tokens/sec/user<span class="o">)</span>:<span class="w"> </span><span class="m">5</span>.4656
|
||
Average<span class="w"> </span>time-to-first-token<span class="w"> </span><span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span><span class="o">(</span>ms<span class="o">)</span>:<span class="w"> </span><span class="m">52667</span>.0339
|
||
Average<span class="w"> </span>time-per-output-token<span class="w"> </span><span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span><span class="o">(</span>ms<span class="o">)</span>:<span class="w"> </span><span class="m">182</span>.9639
|
||
|
||
--<span class="w"> </span>Per-Request<span class="w"> </span>Time-per-Output-Token<span class="w"> </span><span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>Breakdown<span class="w"> </span><span class="o">(</span>ms<span class="o">)</span>
|
||
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>MINIMUM:<span class="w"> </span><span class="m">32</span>.8005
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>MAXIMUM:<span class="w"> </span><span class="m">208</span>.4667
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>AVERAGE:<span class="w"> </span><span class="m">182</span>.9639
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>P50<span class="w"> </span>:<span class="w"> </span><span class="m">204</span>.0463
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>P90<span class="w"> </span>:<span class="w"> </span><span class="m">206</span>.3863
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>P95<span class="w"> </span>:<span class="w"> </span><span class="m">206</span>.5064
|
||
<span class="o">[</span>TPOT<span class="o">]</span><span class="w"> </span>P99<span class="w"> </span>:<span class="w"> </span><span class="m">206</span>.5821
|
||
|
||
--<span class="w"> </span>Per-Request<span class="w"> </span>Time-to-First-Token<span class="w"> </span><span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>Breakdown<span class="w"> </span><span class="o">(</span>ms<span class="o">)</span>
|
||
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>MINIMUM:<span class="w"> </span><span class="m">3914</span>.7621
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>MAXIMUM:<span class="w"> </span><span class="m">107501</span>.2487
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>AVERAGE:<span class="w"> </span><span class="m">52667</span>.0339
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>P50<span class="w"> </span>:<span class="w"> </span><span class="m">52269</span>.7072
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>P90<span class="w"> </span>:<span class="w"> </span><span class="m">96583</span>.7187
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>P95<span class="w"> </span>:<span class="w"> </span><span class="m">101978</span>.4566
|
||
<span class="o">[</span>TTFT<span class="o">]</span><span class="w"> </span>P99<span class="w"> </span>:<span class="w"> </span><span class="m">106563</span>.4497
|
||
|
||
--<span class="w"> </span>Request<span class="w"> </span>Latency<span class="w"> </span>Breakdown<span class="w"> </span><span class="o">(</span>ms<span class="o">)</span><span class="w"> </span>-----------------------
|
||
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>P50<span class="w"> </span>:<span class="w"> </span><span class="m">78509</span>.2102
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>P90<span class="w"> </span>:<span class="w"> </span><span class="m">110804</span>.0017
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>P95<span class="w"> </span>:<span class="w"> </span><span class="m">111302</span>.9101
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>P99<span class="w"> </span>:<span class="w"> </span><span class="m">111618</span>.2158
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>MINIMUM:<span class="w"> </span><span class="m">24189</span>.0838
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>MAXIMUM:<span class="w"> </span><span class="m">111668</span>.0964
|
||
<span class="o">[</span>Latency<span class="o">]</span><span class="w"> </span>AVERAGE:<span class="w"> </span><span class="m">75903</span>.4456
|
||
</pre></div>
|
||
</div>
|
||
<blockquote>
|
||
<div><p>[!WARNING] In some cases, the benchmarker may not print anything at all. This behavior usually
|
||
means that the benchmark has hit an out of memory issue. Try reducing the KV cache percentage
|
||
using the <code class="docutils literal notranslate"><span class="pre">--kv_cache_free_gpu_mem_fraction</span></code> option to lower the percentage of used memory.</p>
|
||
</div></blockquote>
|
||
</section>
|
||
</section>
|
||
</section>
|
||
|
||
|
||
</article>
|
||
|
||
|
||
|
||
|
||
|
||
<footer class="prev-next-footer d-print-none">
|
||
|
||
<div class="prev-next-area">
|
||
</div>
|
||
</footer>
|
||
|
||
</div>
|
||
|
||
|
||
|
||
|
||
|
||
<dialog id="pst-secondary-sidebar-modal"></dialog>
|
||
<div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
|
||
|
||
|
||
<div class="sidebar-secondary-item">
|
||
<div
|
||
id="pst-page-navigation-heading-2"
|
||
class="page-toc tocsection onthispage">
|
||
<i class="fa-solid fa-list"></i> On this page
|
||
</div>
|
||
<nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
|
||
<ul class="visible nav section-nav flex-column">
|
||
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#throughput-measurements">Throughput Measurements</a><ul class="nav section-nav flex-column">
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#hardware">Hardware</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#fp4-models">FP4 Models</a><ul class="nav section-nav flex-column">
|
||
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-3-3-70b-fp4">Llama 3.3 70B FP4</a></li>
|
||
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-3-1-405b-fp4">Llama 3.1 405B FP4</a></li>
|
||
</ul>
|
||
</li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#fp8-models">FP8 Models</a><ul class="nav section-nav flex-column">
|
||
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-3-1-8b-fp8">Llama 3.1 8B FP8</a></li>
|
||
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-3-3-70b-fp8">Llama 3.3 70B FP8</a></li>
|
||
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-3-1-405b-fp8">Llama 3.1 405B FP8</a></li>
|
||
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-4-maverick-fp8">Llama 4 Maverick FP8</a></li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
</li>
|
||
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#reproducing-benchmarked-results">Reproducing Benchmarked Results</a><ul class="nav section-nav flex-column">
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#command-overview">Command Overview</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#variables">Variables</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#preparing-a-dataset">Preparing a Dataset</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#running-the-benchmark">Running the Benchmark</a></li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
</nav></div>
|
||
|
||
</div></div>
|
||
|
||
|
||
|
||
</div>
|
||
<footer class="bd-footer-content">
|
||
|
||
</footer>
|
||
|
||
</main>
|
||
</div>
|
||
</div>
|
||
|
||
<!-- Scripts loaded after <body> so the DOM is not blocked -->
|
||
<script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
|
||
<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
|
||
|
||
<footer class="bd-footer">
|
||
<div class="bd-footer__inner bd-page-width">
|
||
|
||
<div class="footer-items__start">
|
||
|
||
<div class="footer-item">
|
||
<a class="footer-brand logo" href="https://www.nvidia.com">
|
||
<img src="../_static/nvidia-logo-horiz-rgb-1c-blk-for-screen.svg" class="logo__image only-light" alt="NVIDIA"/>
|
||
<img src="../_static/nvidia-logo-horiz-rgb-1c-wht-for-screen.svg" class="logo__image only-dark" alt="NVIDIA"/>
|
||
</a></div>
|
||
|
||
<div class="footer-item">
|
||
|
||
<div class="footer-links">
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-policy/">Privacy Policy</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/about-nvidia/terms-of-service/">Terms of Service</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/about-nvidia/accessibility/">Accessibility</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/about-nvidia/company-policies/">Corporate Policies</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/product-security/">Product Security</a>
|
||
|
|
||
|
||
|
||
|
||
<a class="external" href="https://www.nvidia.com/en-us/contact/">Contact</a>
|
||
|
||
|
||
|
||
</div>
|
||
</div>
|
||
|
||
<div class="footer-item">
|
||
|
||
|
||
|
||
|
||
<p class="copyright">
|
||
|
||
Copyright © 2025, NVidia.
|
||
<br/>
|
||
|
||
</p>
|
||
</div>
|
||
|
||
<div class="footer-item">
|
||
<div class="extra_footer">
|
||
|
||
<p>Last updated on September 15, 2025.</p>
|
||
|
||
<p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/0c9430e">0c9430e</a>.</p>
|
||
|
||
</div></div>
|
||
|
||
</div>
|
||
|
||
|
||
|
||
</div>
|
||
|
||
</footer>
|
||
</body>
|
||
</html> |