kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
03a1561a27
TensorRT-LLMs/architecture/checkpoint.html
Kaiyu Xie 03a1561a27 Update GitHub pages in root to v1.1.0rc3
2025-09-04 03:19:11 +00:00

1049 lines
60 KiB
HTML
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<!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>TensorRT-LLM Checkpoint &#8212; 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 = 'architecture/checkpoint';</script>
<script>
DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.1.0rc3';
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" />
<link rel="next" title="TensorRT-LLM Build Workflow" href="workflow.html" />
<link rel="prev" title="Model Definition" href="core-concepts.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
<meta name="docsearch:version" content="1.1.0rc3" />
</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"><a class="reference internal" href="../key-features.html">Key Features</a></li>
<li class="toctree-l1"><a class="reference internal" href="../torch.html">PyTorch Backend</a></li>
<li class="toctree-l1"><a class="reference internal" href="../release-notes.html">Release Notes</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Installation</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="../installation/containers.html">Pre-built release container images on NGC</a></li>
<li class="toctree-l1"><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-l1"><a class="reference internal" href="../installation/build-from-source-linux.html">Building from Source Code on Linux</a></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"><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 &amp; Hopper Hardware</a></li>
<li class="toctree-l1"><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 &amp; Hopper Hardware</a></li>
<li class="toctree-l1"><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 &amp; Hopper Hardware</a></li>
<li class="toctree-l1"><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>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">LLM API</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">Examples</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1 has-children"><a class="reference internal" href="../examples/index.html">LLM Examples Introduction</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="simple">
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../examples/customization.html">LLM Common Customizations</a></li>
<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>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Model Definition API</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="../python-api/tensorrt_llm.layers.html">Layers</a></li>
<li class="toctree-l1"><a class="reference internal" href="../python-api/tensorrt_llm.functional.html">Functionals</a></li>
<li class="toctree-l1"><a class="reference internal" href="../python-api/tensorrt_llm.models.html">Models</a></li>
<li class="toctree-l1"><a class="reference internal" href="../python-api/tensorrt_llm.plugin.html">Plugin</a></li>
<li class="toctree-l1"><a class="reference internal" href="../python-api/tensorrt_llm.quantization.html">Quantization</a></li>
<li class="toctree-l1"><a class="reference internal" href="../python-api/tensorrt_llm.runtime.html">Runtime</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">C++ API</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="../_cpp_gen/executor.html">Executor</a></li>
<li class="toctree-l1"><a class="reference internal" href="../_cpp_gen/runtime.html">Runtime</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Command-Line 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-build.html">trtllm-build</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">Architecture</span></p>
<ul class="current nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="overview.html">TensorRT-LLM Architecture</a></li>
<li class="toctree-l1"><a class="reference internal" href="core-concepts.html">Model Definition</a></li>
<li class="toctree-l1 current active"><a class="current reference internal" href="#">TensorRT-LLM Checkpoint</a></li>
<li class="toctree-l1"><a class="reference internal" href="workflow.html">TensorRT-LLM Build Workflow</a></li>
<li class="toctree-l1"><a class="reference internal" href="add-model.html">Adding a Model</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Advanced</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="../advanced/gpt-attention.html">Multi-Head, Multi-Query, and Group-Query Attention</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/gpt-runtime.html">C++ GPT Runtime</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/executor.html">Executor API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/graph-rewriting.html">Graph Rewriting Module</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/lora.html">Run gpt-2b + LoRA using Executor / cpp runtime</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/expert-parallelism.html">Expert Parallelism in TensorRT-LLM</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/kv-cache-management.html">KV Cache Management: Pools, Blocks, and Events</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/kv-cache-reuse.html">KV cache reuse</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/speculative-decoding.html">Speculative Sampling</a></li>
<li class="toctree-l1"><a class="reference internal" href="../advanced/disaggregated-service.html">Disaggregated-Service (Prototype)</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Performance</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="../performance/perf-overview.html">Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="../performance/perf-benchmarking.html">Benchmarking</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../performance/performance-tuning-guide/index.html">Performance Tuning Guide</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="../performance/performance-tuning-guide/benchmarking-default-performance.html">Benchmarking Default Performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="../performance/performance-tuning-guide/useful-build-time-flags.html">Useful Build-Time Flags</a></li>
<li class="toctree-l2"><a class="reference internal" href="../performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.html">Tuning Max Batch Size and Max Num Tokens</a></li>
<li class="toctree-l2"><a class="reference internal" href="../performance/performance-tuning-guide/deciding-model-sharding-strategy.html">Deciding Model Sharding Strategy</a></li>
<li class="toctree-l2"><a class="reference internal" href="../performance/performance-tuning-guide/fp8-quantization.html">FP8 Quantization</a></li>
<li class="toctree-l2"><a class="reference internal" href="../performance/performance-tuning-guide/useful-runtime-flags.html">Useful Runtime Options</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../performance/perf-analysis.html">Performance Analysis</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">Reference</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="../reference/troubleshooting.html">Troubleshooting</a></li>
<li class="toctree-l1"><a class="reference internal" href="../reference/support-matrix.html">Support Matrix</a></li>
<li class="toctree-l1"><a class="reference internal" href="../reference/precision.html">Numerical Precision</a></li>
<li class="toctree-l1"><a class="reference internal" href="../reference/memory.html">Memory Usage of TensorRT-LLM</a></li>
<li class="toctree-l1"><a class="reference internal" href="../reference/ci-overview.html">Continuous Integration Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="../reference/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/H100vsA100.html">H100 has 4.6x A100 Performance in TensorRT-LLM, achieving 10,000 tok/s at 100ms to first token</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/Falcon180B-H200.html">Falcon-180B on a single H200 GPU with INT4 AWQ, and 6.7x faster Llama-70B over A100</a></li>
<li class="toctree-l1"><a class="reference internal" href="../blogs/quantization-in-TRT-LLM.html">Speed up inference with SOTA quantization techniques in TRT-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/tech_blog/blog10_ADP_Balance_Strategy.html">ADP Balance Strategy</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>
</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">TensorRT-LLM Checkpoint</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="tensorrt-llm-checkpoint">
<h1>TensorRT-LLM Checkpoint<a class="headerlink" href="#tensorrt-llm-checkpoint" title="Link to this heading">#</a></h1>
<section id="overview">
<h2>Overview<a class="headerlink" href="#overview" title="Link to this heading">#</a></h2>
<p>The earlier versions (pre-0.8 version) of TensorRT-LLM were developed with a very aggressive timeline. For those versions, emphasis was not put on defining a unified workflow. Now that TensorRT-LLM has reached some level of feature richness, the development team has decided to put more effort into unifying the APIs and workflow of TensorRT-LLM. This file documents the workflow around TensorRT-LLM checkpoint and the set of CLI tools to generate checkpoint, build engines, and evaluate engines.</p>
<p>There are three steps in the workflow:</p>
<ol class="arabic simple">
<li><p>Convert weights from different source frameworks into TensorRT-LLM checkpoint.</p></li>
<li><p>Build the TensorRT-LLM checkpoint into TensorRT engines with a unified build command.</p></li>
<li><p>Load the engines to TensorRT-LLM model runner and evaluate with different evaluation tasks.</p></li>
</ol>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">NeMo</span> <span class="o">-------------</span>
<span class="o">|</span>
<span class="n">HuggingFace</span> <span class="o">------</span>
<span class="o">|</span> <span class="n">convert</span> <span class="n">build</span> <span class="n">load</span>
<span class="n">Modelopt</span> <span class="o">---------</span> <span class="o">----------&gt;</span> <span class="n">TensorRT</span><span class="o">-</span><span class="n">LLM</span> <span class="n">Checkpoint</span> <span class="o">--------&gt;</span> <span class="n">TensorRT</span> <span class="n">Engine</span> <span class="o">------&gt;</span> <span class="n">TensorRT</span><span class="o">-</span><span class="n">LLM</span> <span class="n">ModelRunner</span>
<span class="o">|</span>
<span class="n">JAX</span> <span class="o">--------------</span>
<span class="o">|</span>
<span class="n">DeepSpeed</span> <span class="o">--------</span>
</pre></div>
</div>
</section>
<section id="prepare-the-tensorrt-llm-checkpoint">
<h2>Prepare the TensorRT-LLM Checkpoint<a class="headerlink" href="#prepare-the-tensorrt-llm-checkpoint" title="Link to this heading">#</a></h2>
<p>TensorRT-LLM aims at supporting different sources:</p>
<ol class="arabic simple">
<li><p>Trained models from NVIDIA NeMo, Microsoft DeepSpeed, and JAX</p></li>
<li><p>Quantized models from NVIDIA Modelopt</p></li>
<li><p>Popular models from HuggingFace</p></li>
</ol>
<p>TensorRT-LLM defines its own checkpoint format. A checkpoint directory includes:</p>
<ol class="arabic simple">
<li><p>One config <code class="docutils literal notranslate"><span class="pre">json</span></code> file, which contains several model hyper-parameters.</p></li>
<li><p>One or several rank weights files, each file contains a dictionary of tensors (weights).
The different files are loaded by different ranks in a multi-GPU (multi-process) scenario.</p></li>
</ol>
<section id="config">
<h3>Config<a class="headerlink" href="#config" 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>Field</p></th>
<th class="head text-left"><p>Type</p></th>
<th class="head text-left"><p>Default Value</p></th>
</tr>
</thead>
<tbody>
<tr class="row-even"><td class="text-left"><p>architecture</p></td>
<td class="text-left"><p>string</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>dtype</p></td>
<td class="text-left"><p>string</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>logits_dtype</p></td>
<td class="text-left"><p>string</p></td>
<td class="text-left"><p>‘float32’</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>vocab_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>max_position_embeddings</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>null</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>hidden_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>num_hidden_layers</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>num_attention_heads</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>num_key_value_heads</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>num_attention_heads</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>hidden_act</p></td>
<td class="text-left"><p>string</p></td>
<td class="text-left"><p>mandatory</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>intermediate_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>null</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>norm_epsilon</p></td>
<td class="text-left"><p>float</p></td>
<td class="text-left"><p>1e-5</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>position_embedding_type</p></td>
<td class="text-left"><p>string</p></td>
<td class="text-left"><p>‘learned_absolute’</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>mapping.world_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>1</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>mapping.tp_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>1</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>mapping.pp_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>1</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>quantization.quant_algo</p></td>
<td class="text-left"><p>str</p></td>
<td class="text-left"><p>null</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>quantization.kv_cache_quant_algo</p></td>
<td class="text-left"><p>str</p></td>
<td class="text-left"><p>null</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>quantization.group_size</p></td>
<td class="text-left"><p>int</p></td>
<td class="text-left"><p>64</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>quantization.has_zero_point</p></td>
<td class="text-left"><p>bool</p></td>
<td class="text-left"><p>False</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>quantization.pre_quant_scale</p></td>
<td class="text-left"><p>bool</p></td>
<td class="text-left"><p>False</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>quantization.exclude_modules</p></td>
<td class="text-left"><p>list</p></td>
<td class="text-left"><p>null</p></td>
</tr>
</tbody>
</table>
</div>
<p><code class="docutils literal notranslate"><span class="pre">mapping.world_size</span></code> means <code class="docutils literal notranslate"><span class="pre">mapping</span></code> is a dictionary containing the <code class="docutils literal notranslate"><span class="pre">world_size</span></code> sub field.</p>
<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;architecture&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;OPTForCausalLM&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;mapping&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;world_size&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="p">}</span>
<span class="p">}</span>
</pre></div>
</div>
<p>Supported quantization algorithm list:</p>
<ul class="simple">
<li><p>W8A16</p></li>
<li><p>W4A16</p></li>
<li><p>W4A16_AWQ</p></li>
<li><p>W4A8_AWQ</p></li>
<li><p>W4A16_GPTQ</p></li>
<li><p>FP8</p></li>
<li><p>W8A8_SQ_PER_CHANNEL</p></li>
</ul>
<p>Supported KV cache quantization algorithm list:</p>
<ul class="simple">
<li><p>FP8</p></li>
<li><p>INT8</p></li>
</ul>
<p>The config field is extensible, a model could add its own specific config fields.
For example, OPT model has a <code class="docutils literal notranslate"><span class="pre">do_layer_norm_before</span></code> field.</p>
<p>Here is the model specific config list:</p>
<div class="pst-scrollable-table-container"><table class="table">
<thead>
<tr class="row-odd"><th class="head text-left"><p>Field</p></th>
<th class="head text-left"><p>Type</p></th>
<th class="head text-left"><p>Default Value</p></th>
</tr>
</thead>
<tbody>
<tr class="row-even"><td class="text-left"><p>OPT</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>do_layer_norm_before</p></td>
<td class="text-left"><p>bool</p></td>
<td class="text-left"><p>False</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>
</tr>
<tr class="row-odd"><td class="text-left"><p>Falcon</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>bias</p></td>
<td class="text-left"><p>bool</p></td>
<td class="text-left"><p>True</p></td>
</tr>
<tr class="row-odd"><td class="text-left"><p>new_decoder_architecture</p></td>
<td class="text-left"><p>bool</p></td>
<td class="text-left"><p>False</p></td>
</tr>
<tr class="row-even"><td class="text-left"><p>parallel_attention</p></td>
<td class="text-left"><p>bool</p></td>
<td class="text-left"><p>False</p></td>
</tr>
</tbody>
</table>
</div>
</section>
<section id="rank-weights">
<h3>Rank Weights<a class="headerlink" href="#rank-weights" title="Link to this heading">#</a></h3>
<p>Like PyTorch, the tensor (weight) name is a string containing hierarchical information,
which is uniquely mapped to a certain parameter of a TensorRT-LLM model.</p>
<p>For example, each transformer layer of the OPT model contains an <code class="docutils literal notranslate"><span class="pre">Attention</span></code> layer, an <code class="docutils literal notranslate"><span class="pre">MLP</span></code> layer. and two <code class="docutils literal notranslate"><span class="pre">LayerNorm</span></code> layers.</p>
<section id="attention-weights">
<h4>Attention Weights<a class="headerlink" href="#attention-weights" title="Link to this heading">#</a></h4>
<p>The <code class="docutils literal notranslate"><span class="pre">Attention</span></code> layer contains two <code class="docutils literal notranslate"><span class="pre">Linear</span></code> layers, qkv and dense; each <code class="docutils literal notranslate"><span class="pre">Linear</span></code> layer contains one weight and one bias.
There are four tensors (weights) in total, whose names are:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.qkv.weight</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.qkv.bias</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.dense.weight</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.dense.bias</span></code></p></li>
</ul>
<p>where <code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention</span></code> is the prefix name, indicating that the weights/biases are in the Attention module of the 0-th transformer layer.</p>
</section>
<section id="mlp-weights">
<h4>MLP Weights<a class="headerlink" href="#mlp-weights" title="Link to this heading">#</a></h4>
<p>The <code class="docutils literal notranslate"><span class="pre">MLP</span></code> layer also contains two <code class="docutils literal notranslate"><span class="pre">Linear</span></code> layers, fc and proj; each <code class="docutils literal notranslate"><span class="pre">Linear</span></code> layer contains one weight and one bias.
There are four tensors (weights) in total, whose names are:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.fc.weight</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.fc.bias</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.proj.weight</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.proj.bias</span></code></p></li>
</ul>
<p>where <code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp</span></code> is the prefix name, indicating that the weights/biases are in the MLP module of the 0-th transformer layer.</p>
</section>
<section id="layernorm-weights">
<h4>LayerNorm Weights<a class="headerlink" href="#layernorm-weights" title="Link to this heading">#</a></h4>
<p>Each of the two <code class="docutils literal notranslate"><span class="pre">LayerNorm</span></code> layers, namely <code class="docutils literal notranslate"><span class="pre">input_layernorm</span></code> and <code class="docutils literal notranslate"><span class="pre">post_layernorm</span></code>, contains one weight and one bias.
There are four tensors (weights) in total, whose names are:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.input_layernorm.weight</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.input_layernorm.bias</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.post_layernorm.weight</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.post_layernorm.bias</span></code></p></li>
</ul>
<p>where <code class="docutils literal notranslate"><span class="pre">transformer.layers.0.input_layernorm</span></code> and <code class="docutils literal notranslate"><span class="pre">transformer.layers.0.post_layernorm</span></code> are prefix names for the two <code class="docutils literal notranslate"><span class="pre">layernorm</span></code> modules.</p>
</section>
<section id="kv-cache-quantization-scaling-factors">
<h4>KV Cache Quantization Scaling Factors<a class="headerlink" href="#kv-cache-quantization-scaling-factors" title="Link to this heading">#</a></h4>
<p>If we quantize the model, there will be different tensors (depending on the quantization method applied).
For example, if we quantize the KV cache, the <code class="docutils literal notranslate"><span class="pre">Attention</span></code> layer will have this extra scaling factor:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.kv_cache_scaling_factor</span></code></p></li>
</ul>
</section>
<section id="fp8-quantization-scaling-factors">
<h4>FP8 Quantization Scaling Factors<a class="headerlink" href="#fp8-quantization-scaling-factors" title="Link to this heading">#</a></h4>
<p>Here is the FP8 scaling factors of <code class="docutils literal notranslate"><span class="pre">attention.qkv</span></code> linear layer:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.qkv.activation_scaling_factor</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.qkv.weights_scaling_factor</span></code></p></li>
</ul>
</section>
<section id="awq-quantization-scaling-factors">
<h4>AWQ Quantization Scaling Factors<a class="headerlink" href="#awq-quantization-scaling-factors" title="Link to this heading">#</a></h4>
<p>Here is the AWQ scaling factors of <code class="docutils literal notranslate"><span class="pre">mlp.fc</span></code> linear layer:</p>
<ul>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.fc.weights_scaling_factor</span></code></p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.fc.prequant_scaling_factor</span></code></p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>The linear weights in TensorRT-LLM checkpoint always follows (<code class="docutils literal notranslate"><span class="pre">out_feature</span></code>, <code class="docutils literal notranslate"><span class="pre">in_feature</span></code>) shape, whereas some quantized linear in TensorRT-LLM implemented by plugin may use (<code class="docutils literal notranslate"><span class="pre">in_feature</span></code>, <code class="docutils literal notranslate"><span class="pre">out_fature</span></code>) shape. The <code class="docutils literal notranslate"><span class="pre">trtllm-build</span></code> command adds a transpose operation to post-process it.</p>
</div>
</li>
</ul>
</section>
</section>
<section id="example">
<h3>Example<a class="headerlink" href="#example" title="Link to this heading">#</a></h3>
<p>Let’s take OPT as an example and deploy the model with tensor parallelism 2:</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>examples/opt
python3<span class="w"> </span>convert_checkpoint.py<span class="w"> </span>--model_dir<span class="w"> </span>./opt-125m<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--dtype<span class="w"> </span>float16<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--tp_size<span class="w"> </span><span class="m">2</span><span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--output_dir<span class="w"> </span>./opt/125M/trt_ckpt/fp16/2-gpu/
</pre></div>
</div>
<p>Here is the checkpoint directory:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">./</span><span class="n">opt</span><span class="o">/</span><span class="mi">125</span><span class="n">M</span><span class="o">/</span><span class="n">trt_ckpt</span><span class="o">/</span><span class="n">fp16</span><span class="o">/</span><span class="mi">1</span><span class="o">-</span><span class="n">gpu</span><span class="o">/</span>
<span class="n">config</span><span class="o">.</span><span class="n">json</span>
<span class="n">rank0</span><span class="o">.</span><span class="n">safetensors</span>
<span class="n">rank1</span><span class="o">.</span><span class="n">safetensors</span>
</pre></div>
</div>
<p>Here is the <code class="docutils literal notranslate"><span class="pre">config.json</span></code>:</p>
<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;architecture&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;OPTForCausalLM&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;dtype&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;float16&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;logits_dtype&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;float32&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;num_hidden_layers&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">12</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;num_attention_heads&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">12</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;hidden_size&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">768</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;vocab_size&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">50272</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;position_embedding_type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;learned_absolute&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;max_position_embeddings&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">2048</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;hidden_act&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;relu&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;mapping&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;world_size&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">2</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;tp_size&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">2</span>
<span class="w"> </span><span class="p">},</span>
<span class="w"> </span><span class="nt">&quot;use_parallel_embedding&quot;</span><span class="p">:</span><span class="w"> </span><span class="kc">false</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;embedding_sharding_dim&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">0</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;do_layer_norm_before&quot;</span><span class="p">:</span><span class="w"> </span><span class="kc">true</span><span class="p">,</span>
<span class="p">}</span>
</pre></div>
</div>
</section>
</section>
<section id="build-checkpoint-into-tensorrt-engine">
<h2>Build Checkpoint into TensorRT Engine<a class="headerlink" href="#build-checkpoint-into-tensorrt-engine" title="Link to this heading">#</a></h2>
<p>TensorRT-LLM provides a unified build command: <code class="docutils literal notranslate"><span class="pre">trtllm-build</span></code>. Before using it,
you may need to add it to the <code class="docutils literal notranslate"><span class="pre">PATH</span></code>.</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="nb">export</span><span class="w"> </span><span class="nv">PATH</span><span class="o">=</span>/usr/local/bin:<span class="nv">$PATH</span>
trtllm-build<span class="w"> </span>--checkpoint_dir<span class="w"> </span>./opt/125M/trt_ckpt/fp16/2-gpu/<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--gemm_plugin<span class="w"> </span>float16<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--max_batch_size<span class="w"> </span><span class="m">8</span><span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--max_input_len<span class="w"> </span><span class="m">924</span><span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--max_seq_len<span class="w"> </span><span class="m">1024</span><span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--output_dir<span class="w"> </span>./opt/125M/trt_engines/fp16/2-gpu/
</pre></div>
</div>
</section>
<section id="make-evaluation">
<h2>Make Evaluation<a class="headerlink" href="#make-evaluation" title="Link to this heading">#</a></h2>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>mpirun<span class="w"> </span>-n<span class="w"> </span><span class="m">2</span><span class="w"> </span>--allow-run-as-root<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>python3<span class="w"> </span>../summarize.py<span class="w"> </span>--engine_dir<span class="w"> </span>./opt/125M/trt_engines/fp16/2-gpu/<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--batch_size<span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--test_trt_llm<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--hf_model_dir<span class="w"> </span>opt-125m<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--data_type<span class="w"> </span>fp16<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--check_accuracy<span class="w"> </span><span class="se">\</span>
<span class="w"> </span>--tensorrt_llm_rouge1_threshold<span class="o">=</span><span class="m">14</span>
</pre></div>
</div>
</section>
</section>
</article>
<footer class="prev-next-footer d-print-none">
<div class="prev-next-area">
<a class="left-prev"
href="core-concepts.html"
title="previous page">
<i class="fa-solid fa-angle-left"></i>
<div class="prev-next-info">
<p class="prev-next-subtitle">previous</p>
<p class="prev-next-title">Model Definition</p>
</div>
</a>
<a class="right-next"
href="workflow.html"
title="next page">
<div class="prev-next-info">
<p class="prev-next-subtitle">next</p>
<p class="prev-next-title">TensorRT-LLM Build Workflow</p>
</div>
<i class="fa-solid fa-angle-right"></i>
</a>
</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="#overview">Overview</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#prepare-the-tensorrt-llm-checkpoint">Prepare the TensorRT-LLM Checkpoint</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#config">Config</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#rank-weights">Rank Weights</a><ul class="nav section-nav flex-column">
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#attention-weights">Attention Weights</a></li>
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#mlp-weights">MLP Weights</a></li>
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#layernorm-weights">LayerNorm Weights</a></li>
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#kv-cache-quantization-scaling-factors">KV Cache Quantization Scaling Factors</a></li>
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#fp8-quantization-scaling-factors">FP8 Quantization Scaling Factors</a></li>
<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#awq-quantization-scaling-factors">AWQ Quantization Scaling Factors</a></li>
</ul>
</li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#example">Example</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#build-checkpoint-into-tensorrt-engine">Build Checkpoint into TensorRT Engine</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#make-evaluation">Make Evaluation</a></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 02, 2025.</p>
<p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/e81c50d">e81c50d</a>.</p>
</div></div>
</div>
</div>
</footer>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink