mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-22 03:35:00 +08:00
941 lines
67 KiB
HTML
941 lines
67 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>TensorRT-LLM Model Weights Loader — 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/model-weights-loader';</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="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">TensorRT-LLM Model Weights Loader</span></li>
|
||
</ul>
|
||
</nav>
|
||
</div>
|
||
|
||
</div>
|
||
|
||
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
<div id="searchbox"></div>
|
||
<article class="bd-article">
|
||
|
||
<section id="tensorrt-llm-model-weights-loader">
|
||
<h1>TensorRT-LLM Model Weights Loader<a class="headerlink" href="#tensorrt-llm-model-weights-loader" 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 weights loader is designed for easily converting and loading external weight checkpoints into TensorRT-LLM models.</p>
|
||
</section>
|
||
<section id="workflow">
|
||
<h2>Workflow<a class="headerlink" href="#workflow" title="Link to this heading">#</a></h2>
|
||
<p>Weight checkpoints can be generated from all sources, and may have different naming and data layouts compared to TRT-LLM’s requirements. E.g.:</p>
|
||
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># HuggingFace LLaMA checkpoints</span>
|
||
<span class="o">{</span>
|
||
<span class="w"> </span><span class="s2">"model.embed_tokens.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>vocab_size,<span class="w"> </span>hidden_size<span class="o">])</span>
|
||
<span class="w"> </span><span class="s2">"model.layers.0.input_layernorm.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.mlp.down_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>inter_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.mlp.gate_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>inter_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.mlp.up_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>inter_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.post_attention_layernorm.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.self_attn.q_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.self_attn.k_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.self_attn.v_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"model.layers.0.self_attn.o_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span>...,
|
||
<span class="o">}</span>
|
||
<span class="c1"># TensorRT-LLM expected weights</span>
|
||
<span class="o">{</span>
|
||
<span class="w"> </span><span class="s2">"transformer.vocab_embedding.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>vocab_size,<span class="w"> </span>hidden_size<span class="o">])</span>
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.input_layernorm.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.mlp.down_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>inter_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.mlp.gate_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>inter_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.mlp.up_proj.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>inter_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.post_layernorm.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.attention.qkv.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size<span class="w"> </span>*<span class="w"> </span><span class="m">3</span>,<span class="w"> </span>hidden_size<span class="o">])</span>,<span class="w"> </span><span class="c1"># Different layout</span>
|
||
<span class="w"> </span><span class="s2">"transformer.layers.0.attention.dense.weight"</span>:<span class="w"> </span>torch.Tensor<span class="o">([</span>hidden_size,<span class="w"> </span>hidden_size<span class="o">])</span>,
|
||
<span class="w"> </span>...,
|
||
<span class="o">}</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>Conversion means converting the dictionary of <code class="docutils literal notranslate"><span class="pre">{external_keys:external_weights}</span></code> into <code class="docutils literal notranslate"><span class="pre">{tllm_keys:tllm_weights}</span></code>, it includes changing the naming logic and data layouts, and is contains of the following parts:</p>
|
||
<ol class="arabic simple">
|
||
<li><p>Translate a TRT-LLM parameter name into external-format name(s).</p></li>
|
||
<li><p>Loading tensor slice(s) according to the translated names.</p></li>
|
||
<li><p>Postprocess the tensor(s) into target layout.</p></li>
|
||
</ol>
|
||
<section id="translator">
|
||
<h3>Translator<a class="headerlink" href="#translator" title="Link to this heading">#</a></h3>
|
||
<p>TRT-LLM parameter names are translated in units of sections divided by dots. E.g.:</p>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-center"><p>TensorRT-LLM key</p></th>
|
||
<th class="head text-center"><p><code class="docutils literal notranslate"><span class="pre">transformer</span></code></p></th>
|
||
<th class="head"><p>.</p></th>
|
||
<th class="head text-center"><p><code class="docutils literal notranslate"><span class="pre">layers</span></code></p></th>
|
||
<th class="head"><p>.</p></th>
|
||
<th class="head text-center"><p><code class="docutils literal notranslate"><span class="pre">0</span></code></p></th>
|
||
<th class="head"><p>.</p></th>
|
||
<th class="head text-center"><p><code class="docutils literal notranslate"><span class="pre">attention</span></code></p></th>
|
||
<th class="head"><p>.</p></th>
|
||
<th class="head text-center"><p><code class="docutils literal notranslate"><span class="pre">dense</span></code></p></th>
|
||
<th class="head"><p>.</p></th>
|
||
<th class="head text-center"><p><code class="docutils literal notranslate"><span class="pre">weight</span></code></p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-center"><p>Translated external key</p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">model</span></code></p></td>
|
||
<td><p>.</p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">layers</span></code></p></td>
|
||
<td><p>.</p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">0</span></code></p></td>
|
||
<td><p>.</p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">self_attn</span></code></p></td>
|
||
<td><p>.</p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">o_proj</span></code></p></td>
|
||
<td><p>.</p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">weight</span></code></p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
<p>The mapping between TRT-LLM keywords and HF keywords are described in <code class="docutils literal notranslate"><span class="pre">tllm_to_externel_key_dict</span></code> of <code class="docutils literal notranslate"><span class="pre">ModelWeightsLoader</span></code> class object. <br />
|
||
If any of the mappings has one-to-multiple corresponding, the translated key will get multiplied accordingly. E.g.:</p>
|
||
<div class="pst-scrollable-table-container"><table class="table">
|
||
<thead>
|
||
<tr class="row-odd"><th class="head text-center"><p>TensorRT-LLM key and related keyword mapping</p></th>
|
||
<th class="head text-center"><p>Translated external keys</p></th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr class="row-even"><td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.attention.qkv.weight</span></code><br><code class="docutils literal notranslate"><span class="pre">{"qkv":[q_proj,</span> <span class="pre">k_proj,</span> <span class="pre">v_proj]}</span></code></p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">model.layers.0.self_attn.q_proj.weights</span></code><br><code class="docutils literal notranslate"><span class="pre">model.layers.0.self_attn.k_proj.weights</span></code><br><code class="docutils literal notranslate"><span class="pre">model.layers.0.self_attn.v_proj.weights</span></code></p></td>
|
||
</tr>
|
||
<tr class="row-odd"><td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">transformer.layers.0.mlp.fc.weight</span></code><br><code class="docutils literal notranslate"><span class="pre">{"weight":[qweight,</span> <span class="pre">qzeros,</span> <span class="pre">scales]}</span></code></p></td>
|
||
<td class="text-center"><p><code class="docutils literal notranslate"><span class="pre">model.layers.0.mlp.gate_proj.qweight</span></code><br><code class="docutils literal notranslate"><span class="pre">model.layers.0.mlp.gate_proj.qzeros</span></code><br><code class="docutils literal notranslate"><span class="pre">model.layers.0.mlp.gate_proj.scales</span></code></p></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
<p>The default <code class="docutils literal notranslate"><span class="pre">tllm_to_externel_key_dict</span></code> is based on HF LLaMA as:</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">class</span><span class="w"> </span><span class="nc">ModelWeightsLoader</span><span class="p">:</span>
|
||
<span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_dir</span><span class="p">,</span> <span class="n">customized_key_dict</span><span class="p">:</span> <span class="nb">dict</span> <span class="o">=</span> <span class="p">{})</span> <span class="o">-></span> <span class="kc">None</span><span class="p">:</span>
|
||
<span class="o">...</span>
|
||
<span class="bp">self</span><span class="o">.</span><span class="n">tllm_to_externel_key_dict</span> <span class="o">=</span> <span class="p">{</span>
|
||
<span class="s2">"transformer"</span><span class="p">:</span> <span class="s2">"model"</span><span class="p">,</span>
|
||
<span class="s2">"vocab_embedding"</span><span class="p">:</span> <span class="s2">"embed_tokens"</span><span class="p">,</span>
|
||
<span class="s2">"lm_head"</span><span class="p">:</span> <span class="s2">"lm_head"</span><span class="p">,</span>
|
||
<span class="s2">"ln_f"</span><span class="p">:</span> <span class="s2">"norm"</span><span class="p">,</span>
|
||
<span class="s2">"attention"</span><span class="p">:</span> <span class="s2">"self_attn"</span><span class="p">,</span>
|
||
<span class="s2">"qkv"</span><span class="p">:</span> <span class="p">[</span><span class="s2">"q_proj"</span><span class="p">,</span> <span class="s2">"k_proj"</span><span class="p">,</span> <span class="s2">"v_proj"</span><span class="p">],</span>
|
||
<span class="s2">"dense"</span><span class="p">:</span> <span class="s2">"o_proj"</span><span class="p">,</span>
|
||
<span class="s2">"gate"</span><span class="p">:</span> <span class="s2">"up_proj"</span><span class="p">,</span>
|
||
<span class="s2">"proj"</span><span class="p">:</span> <span class="s2">"down_proj"</span><span class="p">,</span>
|
||
<span class="s2">"fc"</span><span class="p">:</span> <span class="s2">"gate_proj"</span><span class="p">,</span>
|
||
<span class="s2">"input_layernorm"</span><span class="p">:</span> <span class="s2">"input_layernorm"</span><span class="p">,</span>
|
||
<span class="s2">"post_layernorm"</span><span class="p">:</span> <span class="s2">"post_attention_layernorm"</span><span class="p">,</span>
|
||
<span class="p">}</span>
|
||
<span class="bp">self</span><span class="o">.</span><span class="n">tllm_to_externel_key_dict</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">customized_key_dict</span><span class="p">)</span>
|
||
<span class="o">...</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>It can be updated through passing <code class="docutils literal notranslate"><span class="pre">customized_key_dict</span></code> when initializing <code class="docutils literal notranslate"><span class="pre">ModelWeightsLoader</span></code>.</p>
|
||
<p>The dictionary will also get updated according to the layer classes. When iterating over parameters,
|
||
if the layer class has attribute <code class="docutils literal notranslate"><span class="pre">tllm_to_externel_key_dict</span></code>, for keywords exist both in the default one and the layer-specified one,
|
||
the weight loader will translate according to the layer attribute with higher priority.
|
||
This can enable the support for different quantization precisions automatically.</p>
|
||
</section>
|
||
<section id="loading-function">
|
||
<h3>Loading function<a class="headerlink" href="#loading-function" title="Link to this heading">#</a></h3>
|
||
<p>The loading function can load an arbitrary tensor slice according to its <code class="docutils literal notranslate"><span class="pre">key</span></code>, <code class="docutils literal notranslate"><span class="pre">tp_size</span></code>, <code class="docutils literal notranslate"><span class="pre">tp_dim</span></code> and <code class="docutils literal notranslate"><span class="pre">tp_rank</span></code>.</p>
|
||
<p>The template for loading function is as following.</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">load_tensor</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">tp_size</span><span class="p">,</span> <span class="n">tp_dim</span><span class="p">,</span> <span class="n">tp_rank</span><span class="p">):</span>
|
||
<span class="c1"># Retrieve file pointer index</span>
|
||
<span class="k">if</span> <span class="n">key</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">shard_map</span><span class="p">:</span>
|
||
<span class="n">ptr_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">shard_map</span><span class="p">[</span><span class="n">key</span><span class="p">]</span>
|
||
<span class="k">else</span><span class="p">:</span>
|
||
<span class="k">return</span> <span class="kc">None</span>
|
||
|
||
<span class="c1"># Load tensor from the corresponding shard</span>
|
||
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">format</span> <span class="o">==</span> <span class="n">ModelWeightsFormat</span><span class="o">.</span><span class="n">SAFETENSORS</span><span class="p">:</span>
|
||
<span class="n">tensor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">shards</span><span class="p">[</span><span class="n">ptr_idx</span><span class="p">]</span><span class="o">.</span><span class="n">get_slice</span><span class="p">(</span><span class="n">key</span><span class="p">)</span>
|
||
<span class="n">tensor_shape</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">get_shape</span><span class="p">()</span>
|
||
<span class="k">else</span><span class="p">:</span>
|
||
<span class="o">...</span>
|
||
|
||
<span class="c1"># Shard and return a tensor slice</span>
|
||
<span class="n">slice_shape</span> <span class="o">=</span> <span class="o">...</span>
|
||
<span class="k">return</span> <span class="n">tensor</span><span class="p">[</span><span class="n">slice_shape</span><span class="p">]</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>When initializing the <code class="docutils literal notranslate"><span class="pre">ModelWeightsLoader</span></code> object, the file format will be derived from <code class="docutils literal notranslate"><span class="pre">model_dir</span></code> through <code class="docutils literal notranslate"><span class="pre">detect_format</span></code>. The following formats are supported for now:</p>
|
||
<ul class="simple">
|
||
<li><p>Directory contains or file named <code class="docutils literal notranslate"><span class="pre">*.safetensors</span></code> (Recommended, has better performance)</p></li>
|
||
<li><p>Directory contains or file named <code class="docutils literal notranslate"><span class="pre">*.bin</span></code></p></li>
|
||
<li><p>Directory contains or file named <code class="docutils literal notranslate"><span class="pre">*.pth</span></code></p></li>
|
||
</ul>
|
||
<p>To support other formats or in-memory loaded models, the format need to be claimed in <code class="docutils literal notranslate"><span class="pre">ModelWeightsFormat</span></code>, <code class="docutils literal notranslate"><span class="pre">detect_format()</span></code>, <code class="docutils literal notranslate"><span class="pre">preload()</span></code> and <code class="docutils literal notranslate"><span class="pre">load_tensor()</span></code>.</p>
|
||
</section>
|
||
<section id="postprocessing-functions">
|
||
<h3>Postprocessing functions<a class="headerlink" href="#postprocessing-functions" title="Link to this heading">#</a></h3>
|
||
<p>After translation and loading, a TRT-LLM key will become a tensor or a list of tensors, which is the input of postprocessing functions. <br />
|
||
Operations including QKV concatenating, MoE weight stacking and weight-only quantization can be handled here.
|
||
The template of postprocessing function is:</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Example for 1-1 weights mapping</span>
|
||
<span class="k">class</span><span class="w"> </span><span class="nc">CustomizedModuleA</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span>
|
||
<span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="o">...</span><span class="p">):</span>
|
||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">...</span><span class="p">)</span>
|
||
<span class="o">...</span>
|
||
<span class="bp">self</span><span class="o">.</span><span class="n">tp_dim</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># Need to set or inherit from parent class</span>
|
||
|
||
<span class="k">def</span><span class="w"> </span><span class="nf">postprocess</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tllm_key</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
|
||
<span class="n">weights</span> <span class="o">=</span> <span class="n">proc</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
|
||
<span class="k">return</span> <span class="p">{</span><span class="n">tllm_key</span><span class="p">:</span> <span class="n">weights</span><span class="p">}</span>
|
||
|
||
<span class="c1"># Example for multiple-multiple weights mapping</span>
|
||
<span class="k">class</span><span class="w"> </span><span class="nc">CustomizedModuleB</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span>
|
||
<span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="o">...</span><span class="p">):</span>
|
||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">...</span><span class="p">)</span>
|
||
<span class="o">...</span>
|
||
<span class="bp">self</span><span class="o">.</span><span class="n">tp_dim</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># Need to set or inherit from parent class</span>
|
||
<span class="c1"># The default value of "weights" in tllm_to_externel_key_dict will be override</span>
|
||
<span class="bp">self</span><span class="o">.</span><span class="n">tllm_to_externel_key_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s2">"weight"</span><span class="p">:</span> <span class="p">[</span><span class="s2">"qweight"</span><span class="p">,</span> <span class="s2">"qzeros"</span><span class="p">,</span> <span class="s2">"scales"</span><span class="p">]}</span>
|
||
|
||
<span class="k">def</span><span class="w"> </span><span class="nf">postprocess</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tllm_key</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
|
||
<span class="c1"># Skipped the postprocess of zeros and weights_scaling_factor</span>
|
||
<span class="c1"># They are loaded in the postprocess of weight</span>
|
||
<span class="n">config</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">"config"</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="c1"># Passed through kwargs by default</span>
|
||
<span class="k">if</span> <span class="ow">not</span> <span class="n">tllm_key</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">"weight"</span><span class="p">):</span>
|
||
<span class="k">return</span> <span class="p">{}</span>
|
||
<span class="c1"># The order in weights is defined in tllm_to_externel_key_dict</span>
|
||
<span class="n">qweight</span><span class="p">,</span> <span class="n">qzeros</span><span class="p">,</span> <span class="n">scales</span> <span class="o">=</span> <span class="n">weights</span>
|
||
<span class="n">proccessed_weight</span><span class="p">,</span> <span class="n">proccessed_zeros</span> <span class="o">=</span> <span class="n">proc</span><span class="p">(</span><span class="n">qweight</span><span class="p">,</span> <span class="n">qzeros</span><span class="p">,</span> <span class="n">config</span><span class="o">.</span><span class="n">num_heads</span><span class="p">)</span>
|
||
<span class="k">return</span> <span class="p">{</span>
|
||
<span class="n">tllm_key</span><span class="p">:</span> <span class="n">proccessed_weight</span><span class="p">,</span>
|
||
<span class="n">tllm_key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"weight"</span><span class="p">,</span> <span class="s2">"zeros"</span><span class="p">):</span> <span class="n">proccessed_zeros</span><span class="p">,</span>
|
||
<span class="n">tllm_key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">"weight"</span><span class="p">,</span> <span class="s2">"weights_scaling_factor"</span><span class="p">):</span> <span class="n">scales</span><span class="p">,</span>
|
||
<span class="p">}</span>
|
||
</pre></div>
|
||
</div>
|
||
</section>
|
||
</section>
|
||
<section id="examples">
|
||
<h2>Examples<a class="headerlink" href="#examples" title="Link to this heading">#</a></h2>
|
||
<p>The <code class="docutils literal notranslate"><span class="pre">ModelWeightsLoader</span></code> class can support different models with the following levels:</p>
|
||
<section id="natively-supported-models">
|
||
<h3>Natively supported models<a class="headerlink" href="#natively-supported-models" title="Link to this heading">#</a></h3>
|
||
<p>For models with native support, users can call the default weight loader without any other operations.</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Using the model weights loader for LLaMA</span>
|
||
<span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm.models.model_weights_loader</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelWeightsLoader</span>
|
||
<span class="n">loader</span> <span class="o">=</span> <span class="n">ModelWeightsLoader</span><span class="p">(</span><span class="n">external_checkpoint_dir</span><span class="p">)</span>
|
||
<span class="n">loader</span><span class="o">.</span><span class="n">generate_tllm_weights</span><span class="p">(</span><span class="n">trtllm_model</span><span class="p">)</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>For calibration-free quantization precisions, passing a properly quantized <code class="docutils literal notranslate"><span class="pre">trtllm_model</span></code> will let the weight loader load at the given precision accordingly. The configurations will be read from <code class="docutils literal notranslate"><span class="pre">trtllm_model.config</span></code> automatically. For now, LLaMA family models using the default <code class="docutils literal notranslate"><span class="pre">tllm_to_externel_key_dict</span></code> is supported natively.</p>
|
||
</section>
|
||
<section id="models-with-customized-key-names">
|
||
<h3>Models with customized key names<a class="headerlink" href="#models-with-customized-key-names" title="Link to this heading">#</a></h3>
|
||
<p>For models with different naming logic, users can still call the default weight loader with <code class="docutils literal notranslate"><span class="pre">customized_key_dict</span></code> specified.</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Using the model weights loader for the LLM part of LLaVA</span>
|
||
<span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm.models.model_weights_loader</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelWeightsLoader</span>
|
||
<span class="n">llava_dict</span> <span class="o">=</span> <span class="p">{</span>
|
||
<span class="s2">"transformer"</span><span class="p">:</span> <span class="s2">"language_model.model"</span><span class="p">,</span>
|
||
<span class="s2">"lm_head"</span><span class="p">:</span> <span class="s2">"language_model.lm_head"</span>
|
||
<span class="p">}</span>
|
||
<span class="n">loader</span> <span class="o">=</span> <span class="n">ModelWeightsLoader</span><span class="p">(</span><span class="n">external_checkpoint_dir</span><span class="p">,</span> <span class="n">llava_dict</span><span class="p">)</span>
|
||
<span class="n">loader</span><span class="o">.</span><span class="n">generate_tllm_weights</span><span class="p">(</span><span class="n">trtllm_model</span><span class="p">)</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>Users need to specify the different part from the default <code class="docutils literal notranslate"><span class="pre">tllm_to_externel_key_dict</span></code>. The loader still have support across different precisions.
|
||
The support for LLaVA and Exaone is in <code class="docutils literal notranslate"><span class="pre">LLaMAForCausalLM.from_hugging_face()</span></code> of <a class="reference download internal" download="" href="../_downloads/408e9af6e2b04a79e78215bde246e8bc/model.py"><span class="xref download myst">model.py</span></a>, and can also be taken as examples.</p>
|
||
</section>
|
||
<section id="models-with-customized-weight-layout">
|
||
<h3>Models with customized weight layout<a class="headerlink" href="#models-with-customized-weight-layout" title="Link to this heading">#</a></h3>
|
||
<p>For models with different weight layout, users can write the conversion loop explicitly and do customized operations.</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Using the model weights loader for BLOOM</span>
|
||
<span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm.models.model_weights_loader</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelWeightsLoader</span>
|
||
<span class="n">bloom_dict</span> <span class="o">=</span> <span class="p">{</span>
|
||
<span class="s2">"transformer"</span><span class="p">:</span> <span class="s2">""</span><span class="p">,</span>
|
||
<span class="s2">"layers"</span><span class="p">:</span> <span class="s2">"h"</span><span class="p">,</span>
|
||
<span class="s2">"ln_f"</span><span class="p">:</span> <span class="s2">"ln_f"</span><span class="p">,</span>
|
||
<span class="s2">"lm_head"</span><span class="p">:</span> <span class="s2">"word_embeddings"</span><span class="p">,</span>
|
||
<span class="s2">"ln_embed"</span><span class="p">:</span> <span class="s2">"word_embeddings_layernorm"</span><span class="p">,</span>
|
||
<span class="s2">"vocab_embedding"</span><span class="p">:</span> <span class="s2">"word_embeddings"</span><span class="p">,</span>
|
||
<span class="s2">"attention"</span><span class="p">:</span> <span class="s2">"self_attention"</span><span class="p">,</span>
|
||
<span class="s2">"qkv"</span><span class="p">:</span> <span class="s2">"query_key_value"</span><span class="p">,</span>
|
||
<span class="s2">"dense"</span><span class="p">:</span> <span class="s2">"dense"</span><span class="p">,</span>
|
||
<span class="s2">"fc"</span><span class="p">:</span> <span class="s2">"dense_h_to_4h"</span><span class="p">,</span>
|
||
<span class="s2">"proj"</span><span class="p">:</span> <span class="s2">"dense_4h_to_h"</span><span class="p">,</span>
|
||
<span class="s2">"post_layernorm"</span><span class="p">:</span> <span class="s2">"post_attention_layernorm"</span><span class="p">,</span>
|
||
<span class="p">}</span>
|
||
<span class="n">loader</span> <span class="o">=</span> <span class="n">ModelWeightsLoader</span><span class="p">(</span><span class="n">external_checkpoint_dir</span><span class="p">,</span> <span class="n">bloom_dict</span><span class="p">)</span>
|
||
<span class="c1"># See ModelWeightsLoader.generate_tllm_weights()</span>
|
||
<span class="n">loader</span><span class="o">.</span><span class="n">update_key_mapping</span><span class="p">(</span><span class="n">trtllm_model</span><span class="p">)</span>
|
||
<span class="n">tllm_weights</span> <span class="o">=</span> <span class="p">{}</span>
|
||
<span class="k">for</span> <span class="n">tllm_key</span><span class="p">,</span> <span class="n">_</span> <span class="ow">in</span> <span class="n">tqdm</span><span class="p">(</span><span class="n">trtllm_model</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()):</span>
|
||
<span class="k">if</span> <span class="n">tllm_key</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">"qkv"</span><span class="p">):</span>
|
||
<span class="c1"># Passing the callable handle</span>
|
||
<span class="n">tllm_weights</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">loader</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">tllm_key</span><span class="p">,</span> <span class="n">preprocess</span><span class="o">=</span><span class="n">customized_preprocess</span><span class="p">))</span>
|
||
<span class="k">else</span><span class="p">:</span>
|
||
<span class="n">tllm_weights</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">loader</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">tllm_key</span><span class="p">))</span>
|
||
<span class="n">loader</span><span class="o">.</span><span class="n">fill</span><span class="p">(</span><span class="n">tllm_weights</span><span class="p">)</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>This will apply <code class="docutils literal notranslate"><span class="pre">preprocess</span></code> after <code class="docutils literal notranslate"><span class="pre">load_tensor()</span></code> and before <code class="docutils literal notranslate"><span class="pre">postprocess</span></code>, and demonstrates how to convert the loaded shard into default HF layout. The loader still have support for precisions quantized from FP16/BF16 (e.g. INT8-wo/INT4-wo), the other precisions may require special operations, and can be addressed inside the <code class="docutils literal notranslate"><span class="pre">preprocess</span></code> function.
|
||
The support for Qwen-1 is in <code class="docutils literal notranslate"><span class="pre">QWenForCausalLM.from_hugging_face()</span></code> of <a class="reference download internal" download="" href="../_downloads/b6815cf245cc7dc7a26a6f727fdc2dc4/model.py"><span class="xref download myst">model.py</span></a>, and can also be taken as example.</p>
|
||
</section>
|
||
<section id="fully-customized">
|
||
<h3>Fully customized<a class="headerlink" href="#fully-customized" title="Link to this heading">#</a></h3>
|
||
<p>If the model weights loader cannot satisfy the requirements, users can write the conversion loop totally on their own.</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">tllm_weights</span> <span class="o">=</span> <span class="p">{}</span>
|
||
<span class="k">for</span> <span class="n">tllm_key</span><span class="p">,</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">tqdm</span><span class="p">(</span><span class="n">trtllm_model</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()):</span>
|
||
<span class="c1"># Load from external checkpoints</span>
|
||
<span class="c1"># The load_tensor() function can also be called here</span>
|
||
<span class="n">tensor</span> <span class="o">=</span> <span class="o">...</span>
|
||
<span class="c1"># Convert tensor and set the values according to the config</span>
|
||
<span class="k">if</span> <span class="n">trtllm_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">quant_algo</span> <span class="o">==</span> <span class="n">xxx</span><span class="p">:</span>
|
||
<span class="o">...</span>
|
||
<span class="k">else</span><span class="p">:</span>
|
||
<span class="o">...</span>
|
||
<span class="n">param</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">tensor</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>In this mode, every precision require user’s own support.</p>
|
||
</section>
|
||
</section>
|
||
<section id="trouble-shooting">
|
||
<h2>Trouble shooting<a class="headerlink" href="#trouble-shooting" title="Link to this heading">#</a></h2>
|
||
<p>The weights loader is enabled for LLaMA family models and Qwen models by default with TensorRT flow only.</p>
|
||
<p>If users are encountered with failure caused by <code class="docutils literal notranslate"><span class="pre">ModelWeightsLoader</span></code>, a workaround is passing environmental variable <code class="docutils literal notranslate"><span class="pre">TRTLLM_DISABLE_UNIFIED_CONVERTER=1</span></code> to disable the model weights loader and fallback to the legacy path.</p>
|
||
<p>This workaround will be removed in future version after the LLaMA/Qwen weights conversion is stable.</p>
|
||
</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="#overview">Overview</a></li>
|
||
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#workflow">Workflow</a><ul class="nav section-nav flex-column">
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#translator">Translator</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#loading-function">Loading function</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#postprocessing-functions">Postprocessing functions</a></li>
|
||
</ul>
|
||
</li>
|
||
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#examples">Examples</a><ul class="nav section-nav flex-column">
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#natively-supported-models">Natively supported models</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#models-with-customized-key-names">Models with customized key names</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#models-with-customized-weight-layout">Models with customized weight layout</a></li>
|
||
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#fully-customized">Fully customized</a></li>
|
||
</ul>
|
||
</li>
|
||
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#trouble-shooting">Trouble shooting</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 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> |