mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-26 21:53:30 +08:00
917 lines
74 KiB
HTML
917 lines
74 KiB
HTML
|
||
|
||
<!DOCTYPE html>
|
||
|
||
|
||
<html lang="en" data-content_root="../" >
|
||
|
||
<head>
|
||
<meta charset="utf-8" />
|
||
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
|
||
|
||
<title>KV Cache Connector — 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 = 'examples/llm_kv_cache_connector';</script>
|
||
<script>
|
||
DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
|
||
DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
|
||
DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.1.0rc2';
|
||
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="Runtime Configuration Examples" href="llm_runtime.html" />
|
||
<link rel="prev" title="Speculative Decoding" href="llm_speculative_decoding.html" />
|
||
|
||
<meta name="viewport" content="width=device-width, initial-scale=1"/>
|
||
<meta name="docsearch:language" content="en"/>
|
||
<meta name="docsearch:version" content="1.1.0rc2" />
|
||
|
||
|
||
</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>
|
||
|
||
|
||
|
||
</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 & 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 & 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 & 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="current nav bd-sidenav">
|
||
<li class="toctree-l1 has-children"><a class="reference internal" href="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="customization.html">LLM Common Customizations</a></li>
|
||
<li class="toctree-l1 current active has-children"><a class="reference internal" href="llm_api_examples.html">LLM Examples</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_inference.html">Generate text</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_inference_async.html">Generate text asynchronously</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_inference_async_streaming.html">Generate text in streaming</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_inference_distributed.html">Distributed LLM Generation</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_guided_decoding.html">Generate text with guided decoding</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_logits_processor.html">Control generated text using logits processor</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_multilora.html">Generate text with multiple LoRA adapters</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_speculative_decoding.html">Speculative Decoding</a></li>
|
||
<li class="toctree-l2 current active"><a class="current reference internal" href="#">KV Cache Connector</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_runtime.html">Runtime Configuration Examples</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_sampling.html">Sampling Techniques Showcase</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_mgmn_llm_distributed.html">Run LLM-API with pytorch backend on Slurm</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="llm_mgmn_trtllm_bench.html">Run trtllm-bench with pytorch backend on Slurm</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="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="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="curl_chat_client.html">Curl Chat Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="curl_chat_client_for_multimodal.html">Curl Chat Client For Multimodal</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="curl_completion_client.html">Curl Completion Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="deepseek_r1_reasoning_parser.html">Deepseek R1 Reasoning Parser</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="genai_perf_client.html">Genai Perf Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="genai_perf_client_for_multimodal.html">Genai Perf Client For Multimodal</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="openai_chat_client.html">OpenAI Chat Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="openai_chat_client_for_multimodal.html">OpenAI Chat Client for Multimodal</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="openai_completion_client.html">OpenAI Completion Client</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="openai_completion_client_for_lora.html">Openai Completion Client For Lora</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="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="nav bd-sidenav">
|
||
<li class="toctree-l1"><a class="reference internal" href="../architecture/overview.html">TensorRT-LLM Architecture</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../architecture/core-concepts.html">Model Definition</a></li>
|
||
|
||
|
||
|
||
<li class="toctree-l1"><a class="reference internal" href="../architecture/checkpoint.html">TensorRT-LLM Checkpoint</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../architecture/workflow.html">TensorRT-LLM Build Workflow</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="../architecture/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>
|
||
</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"><a href="llm_api_examples.html" class="nav-link">LLM Examples</a></li>
|
||
|
||
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">KV Cache Connector</span></li>
|
||
</ul>
|
||
</nav>
|
||
</div>
|
||
|
||
</div>
|
||
|
||
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
<div id="searchbox"></div>
|
||
<article class="bd-article">
|
||
|
||
<section id="kv-cache-connector">
|
||
<h1>KV Cache Connector<a class="headerlink" href="#kv-cache-connector" title="Link to this heading">#</a></h1>
|
||
<p>Source <a class="github reference external" href="https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/llm-api/llm_kv_cache_connector.py">NVIDIA/TensorRT-LLM</a>.</p>
|
||
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="linenos"> 1</span>
|
||
<span class="linenos"> 2</span><span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
|
||
<span class="linenos"> 3</span><span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
|
||
<span class="linenos"> 4</span><span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span><span class="p">,</span> <span class="n">field</span>
|
||
<span class="linenos"> 5</span><span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
|
||
<span class="linenos"> 6</span><span class="kn">from</span><span class="w"> </span><span class="nn">tempfile</span><span class="w"> </span><span class="kn">import</span> <span class="n">TemporaryDirectory</span>
|
||
<span class="linenos"> 7</span>
|
||
<span class="linenos"> 8</span><span class="kn">import</span><span class="w"> </span><span class="nn">click</span>
|
||
<span class="linenos"> 9</span><span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
|
||
<span class="linenos"> 10</span>
|
||
<span class="linenos"> 11</span><span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm</span><span class="w"> </span><span class="kn">import</span> <span class="n">LLM</span><span class="p">,</span> <span class="n">SamplingParams</span><span class="p">,</span> <span class="n">logger</span>
|
||
<span class="linenos"> 12</span><span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm._torch.pyexecutor.kv_cache_connector</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
|
||
<span class="linenos"> 13</span> <span class="n">KvCacheConnectorScheduler</span><span class="p">,</span> <span class="n">KvCacheConnectorWorker</span><span class="p">,</span> <span class="n">SchedulerOutput</span><span class="p">)</span>
|
||
<span class="linenos"> 14</span><span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm.bindings.executor</span><span class="w"> </span><span class="kn">import</span> <span class="n">ExecutorConfig</span>
|
||
<span class="linenos"> 15</span><span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm.bindings.internal.batch_manager</span><span class="w"> </span><span class="kn">import</span> <span class="n">LlmRequest</span>
|
||
<span class="linenos"> 16</span><span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm.llmapi.llm_args</span><span class="w"> </span><span class="kn">import</span> <span class="n">KvCacheConnectorConfig</span>
|
||
<span class="linenos"> 17</span>
|
||
<span class="linenos"> 18</span><span class="c1"># This is a simple example of the use of the KV cache connector.</span>
|
||
<span class="linenos"> 19</span><span class="c1"># It persists KV cache contents into a folder, and can load them back on subsequent runs.</span>
|
||
<span class="linenos"> 20</span><span class="c1"># See tensorrt_llm/_torch/pyexecutor/connector.py for details about the KV cache connector interface.</span>
|
||
<span class="linenos"> 21</span><span class="c1"># NOTE: This example connector implementation is NOT suitable for production use.</span>
|
||
<span class="linenos"> 22</span>
|
||
<span class="linenos"> 23</span><span class="n">CONNECTOR_CACHE_FOLDER_KEY</span> <span class="o">=</span> <span class="s2">"CONNECTOR_CACHE_FOLDER"</span>
|
||
<span class="linenos"> 24</span>
|
||
<span class="linenos"> 25</span>
|
||
<span class="linenos"> 26</span><span class="nd">@dataclass</span>
|
||
<span class="linenos"> 27</span><span class="k">class</span><span class="w"> </span><span class="nc">PersistentKvCacheConnectorMetadata</span><span class="p">:</span>
|
||
<span class="linenos"> 28</span> <span class="n">load</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default_factory</span><span class="o">=</span><span class="nb">list</span><span class="p">)</span>
|
||
<span class="linenos"> 29</span> <span class="n">save</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default_factory</span><span class="o">=</span><span class="nb">list</span><span class="p">)</span>
|
||
<span class="linenos"> 30</span>
|
||
<span class="linenos"> 31</span>
|
||
<span class="linenos"> 32</span><span class="k">class</span><span class="w"> </span><span class="nc">PersistentKvCacheConnectorWorker</span><span class="p">(</span><span class="n">KvCacheConnectorWorker</span><span class="p">):</span>
|
||
<span class="linenos"> 33</span>
|
||
<span class="linenos"> 34</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">executor_config</span><span class="p">:</span> <span class="n">ExecutorConfig</span><span class="p">):</span>
|
||
<span class="linenos"> 35</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="n">executor_config</span><span class="p">)</span>
|
||
<span class="linenos"> 36</span>
|
||
<span class="linenos"> 37</span> <span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_tensor</span> <span class="o">=</span> <span class="kc">None</span>
|
||
<span class="linenos"> 38</span>
|
||
<span class="linenos"> 39</span> <span class="k">def</span><span class="w"> </span><span class="nf">register_kv_caches</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">kv_cache_tensor</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
|
||
<span class="linenos"> 40</span> <span class="k">assert</span> <span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_tensor</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">,</span> <span class="s2">"KV cache tensor already registered"</span>
|
||
<span class="linenos"> 41</span> <span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_tensor</span> <span class="o">=</span> <span class="n">kv_cache_tensor</span>
|
||
<span class="linenos"> 42</span>
|
||
<span class="linenos"> 43</span> <span class="k">def</span><span class="w"> </span><span class="nf">start_load_kv</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">stream</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">Stream</span><span class="p">):</span>
|
||
<span class="linenos"> 44</span> <span class="c1"># Do all loads synchronously, and blockwise.</span>
|
||
<span class="linenos"> 45</span> <span class="k">for</span> <span class="n">path</span><span class="p">,</span> <span class="n">block_id</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_metadata</span><span class="o">.</span><span class="n">load</span><span class="p">:</span>
|
||
<span class="linenos"> 46</span> <span class="n">cpu_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">"cpu"</span><span class="p">)</span>
|
||
<span class="linenos"> 47</span>
|
||
<span class="linenos"> 48</span> <span class="c1"># Copy into the device block.</span>
|
||
<span class="linenos"> 49</span> <span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_tensor</span><span class="p">[</span><span class="n">block_id</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">cpu_tensor</span><span class="p">,</span> <span class="n">non_blocking</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
|
||
<span class="linenos"> 50</span>
|
||
<span class="linenos"> 51</span> <span class="k">def</span><span class="w"> </span><span class="nf">wait_for_layer_load</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">layer_idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stream</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">Stream</span><span class="p">):</span>
|
||
<span class="linenos"> 52</span> <span class="k">pass</span>
|
||
<span class="linenos"> 53</span>
|
||
<span class="linenos"> 54</span> <span class="k">def</span><span class="w"> </span><span class="nf">save_kv_layer</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">layer_idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stream</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">Stream</span><span class="p">):</span>
|
||
<span class="linenos"> 55</span> <span class="k">pass</span>
|
||
<span class="linenos"> 56</span>
|
||
<span class="linenos"> 57</span> <span class="k">def</span><span class="w"> </span><span class="nf">wait_for_save</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">stream</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">Stream</span><span class="p">):</span>
|
||
<span class="linenos"> 58</span>
|
||
<span class="linenos"> 59</span> <span class="c1"># Make sure the forward pass is complete before beginning our save.</span>
|
||
<span class="linenos"> 60</span> <span class="n">stream</span><span class="o">.</span><span class="n">synchronize</span><span class="p">()</span>
|
||
<span class="linenos"> 61</span>
|
||
<span class="linenos"> 62</span> <span class="k">for</span> <span class="n">path</span><span class="p">,</span> <span class="n">block_id</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_metadata</span><span class="o">.</span><span class="n">save</span><span class="p">:</span>
|
||
<span class="linenos"> 63</span> <span class="n">cpu_tensor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_tensor</span><span class="p">[</span><span class="n">block_id</span><span class="p">]</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span>
|
||
<span class="linenos"> 64</span>
|
||
<span class="linenos"> 65</span> <span class="c1"># Don't write anything if this specific block already exists.</span>
|
||
<span class="linenos"> 66</span> <span class="k">if</span> <span class="n">Path</span><span class="p">(</span><span class="n">path</span><span class="p">)</span><span class="o">.</span><span class="n">exists</span><span class="p">():</span>
|
||
<span class="linenos"> 67</span> <span class="k">continue</span>
|
||
<span class="linenos"> 68</span>
|
||
<span class="linenos"> 69</span> <span class="c1"># Do a blocking save to the file. This way, we only return once all saves are complete.</span>
|
||
<span class="linenos"> 70</span> <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">cpu_tensor</span><span class="p">,</span> <span class="n">path</span><span class="p">)</span>
|
||
<span class="linenos"> 71</span>
|
||
<span class="linenos"> 72</span> <span class="k">def</span><span class="w"> </span><span class="nf">get_finished</span><span class="p">(</span>
|
||
<span class="linenos"> 73</span> <span class="bp">self</span><span class="p">,</span> <span class="n">finished_gen_req_ids</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
|
||
<span class="linenos"> 74</span> <span class="n">started_loading_req_ids</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-></span> <span class="nb">tuple</span><span class="p">[</span><span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
|
||
<span class="linenos"> 75</span>
|
||
<span class="linenos"> 76</span> <span class="k">return</span> <span class="p">[],</span> <span class="p">[]</span>
|
||
<span class="linenos"> 77</span>
|
||
<span class="linenos"> 78</span>
|
||
<span class="linenos"> 79</span><span class="k">class</span><span class="w"> </span><span class="nc">PersistentKvCacheConnectorLeader</span><span class="p">(</span><span class="n">KvCacheConnectorScheduler</span><span class="p">):</span>
|
||
<span class="linenos"> 80</span>
|
||
<span class="linenos"> 81</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">executor_config</span><span class="p">:</span> <span class="n">ExecutorConfig</span><span class="p">):</span>
|
||
<span class="linenos"> 82</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="n">executor_config</span><span class="p">)</span>
|
||
<span class="linenos"> 83</span>
|
||
<span class="linenos"> 84</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_config</span><span class="o">.</span><span class="n">tokens_per_block</span>
|
||
<span class="linenos"> 85</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span> <span class="o">=</span> <span class="p">{}</span>
|
||
<span class="linenos"> 86</span>
|
||
<span class="linenos"> 87</span> <span class="bp">self</span><span class="o">.</span><span class="n">cache_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">CONNECTOR_CACHE_FOLDER_KEY</span><span class="p">,</span>
|
||
<span class="linenos"> 88</span> <span class="s2">"./connector_cache"</span><span class="p">)</span>
|
||
<span class="linenos"> 89</span>
|
||
<span class="linenos"> 90</span> <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cache_folder</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
|
||
<span class="linenos"> 91</span>
|
||
<span class="linenos"> 92</span> <span class="k">def</span><span class="w"> </span><span class="nf">build_connector_meta</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">scheduler_output</span><span class="p">:</span> <span class="n">SchedulerOutput</span><span class="p">):</span>
|
||
<span class="linenos"> 93</span> <span class="c1"># NOTE: This is a simplified implementation, and does not work with chunked prefill.</span>
|
||
<span class="linenos"> 94</span>
|
||
<span class="linenos"> 95</span> <span class="n">metadata</span> <span class="o">=</span> <span class="n">PersistentKvCacheConnectorMetadata</span><span class="p">()</span>
|
||
<span class="linenos"> 96</span>
|
||
<span class="linenos"> 97</span> <span class="k">for</span> <span class="n">req</span> <span class="ow">in</span> <span class="n">scheduler_output</span><span class="o">.</span><span class="n">new_requests</span><span class="p">:</span>
|
||
<span class="linenos"> 98</span> <span class="c1"># If we don't have any pending loads for this request, we can skip it.</span>
|
||
<span class="linenos"> 99</span> <span class="k">if</span> <span class="n">req</span><span class="o">.</span><span class="n">request_id</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span><span class="p">:</span>
|
||
<span class="linenos">100</span> <span class="k">continue</span>
|
||
<span class="linenos">101</span>
|
||
<span class="linenos">102</span> <span class="n">num_computed_blocks</span> <span class="o">=</span> <span class="n">req</span><span class="o">.</span><span class="n">computed_position</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span>
|
||
<span class="linenos">103</span> <span class="n">block_ids</span> <span class="o">=</span> <span class="n">req</span><span class="o">.</span><span class="n">new_block_ids</span>
|
||
<span class="linenos">104</span>
|
||
<span class="linenos">105</span> <span class="n">pending_load</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span><span class="p">[</span><span class="n">req</span><span class="o">.</span><span class="n">request_id</span><span class="p">]</span>
|
||
<span class="linenos">106</span>
|
||
<span class="linenos">107</span> <span class="k">for</span> <span class="n">file_path</span><span class="p">,</span> <span class="n">block_pos</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span>
|
||
<span class="linenos">108</span> <span class="n">pending_load</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_computed_blocks</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">block_ids</span><span class="p">))):</span>
|
||
<span class="linenos">109</span> <span class="n">metadata</span><span class="o">.</span><span class="n">load</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">file_path</span><span class="p">,</span> <span class="n">block_ids</span><span class="p">[</span><span class="n">block_pos</span><span class="p">]))</span>
|
||
<span class="linenos">110</span>
|
||
<span class="linenos">111</span> <span class="c1"># Break up the remainder of the token sequence into chunks.</span>
|
||
<span class="linenos">112</span> <span class="n">chunks</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_chunk_tokens</span><span class="p">(</span><span class="n">req</span><span class="o">.</span><span class="n">new_tokens</span><span class="p">)</span>
|
||
<span class="linenos">113</span>
|
||
<span class="linenos">114</span> <span class="c1"># For each chunk that isn't already on device, and isn't in our connector cache, we need to save it.</span>
|
||
<span class="linenos">115</span> <span class="k">for</span> <span class="n">block_pos</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_computed_blocks</span> <span class="o">+</span> <span class="nb">len</span><span class="p">(</span><span class="n">pending_load</span><span class="p">),</span>
|
||
<span class="linenos">116</span> <span class="nb">len</span><span class="p">(</span><span class="n">block_ids</span><span class="p">)):</span>
|
||
<span class="linenos">117</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">chunks</span><span class="p">[</span><span class="n">block_pos</span><span class="p">])</span> <span class="o">==</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">:</span>
|
||
<span class="linenos">118</span> <span class="n">hashed_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_hash_tokens</span><span class="p">(</span><span class="n">chunks</span><span class="p">[</span><span class="n">block_pos</span><span class="p">])</span>
|
||
<span class="linenos">119</span>
|
||
<span class="linenos">120</span> <span class="n">file_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_file_path</span><span class="p">(</span><span class="n">hashed_tokens</span><span class="p">)</span>
|
||
<span class="linenos">121</span>
|
||
<span class="linenos">122</span> <span class="n">metadata</span><span class="o">.</span><span class="n">save</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">file_path</span><span class="p">,</span> <span class="n">block_ids</span><span class="p">[</span><span class="n">block_pos</span><span class="p">]))</span>
|
||
<span class="linenos">123</span>
|
||
<span class="linenos">124</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span> <span class="o">=</span> <span class="p">{}</span>
|
||
<span class="linenos">125</span>
|
||
<span class="linenos">126</span> <span class="k">return</span> <span class="n">metadata</span>
|
||
<span class="linenos">127</span>
|
||
<span class="linenos">128</span> <span class="k">def</span><span class="w"> </span><span class="nf">_hash_tokens</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tokens</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-></span> <span class="nb">int</span><span class="p">:</span>
|
||
<span class="linenos">129</span> <span class="k">return</span> <span class="nb">abs</span><span class="p">(</span><span class="nb">hash</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">tokens</span><span class="p">)))</span>
|
||
<span class="linenos">130</span>
|
||
<span class="linenos">131</span> <span class="k">def</span><span class="w"> </span><span class="nf">_file_path</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">hash_value</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-></span> <span class="n">Path</span><span class="p">:</span>
|
||
<span class="linenos">132</span> <span class="k">return</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cache_folder</span><span class="p">)</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="n">hash_value</span><span class="si">}</span><span class="s2">.pt"</span>
|
||
<span class="linenos">133</span>
|
||
<span class="linenos">134</span> <span class="k">def</span><span class="w"> </span><span class="nf">_chunk_tokens</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tokens</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-></span> <span class="nb">list</span><span class="p">[</span><span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
|
||
<span class="linenos">135</span> <span class="k">return</span> <span class="p">[</span>
|
||
<span class="linenos">136</span> <span class="n">tokens</span><span class="p">[</span><span class="n">i</span><span class="p">:</span><span class="n">i</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">]</span>
|
||
<span class="linenos">137</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokens</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">)</span>
|
||
<span class="linenos">138</span> <span class="p">]</span>
|
||
<span class="linenos">139</span>
|
||
<span class="linenos">140</span> <span class="k">def</span><span class="w"> </span><span class="nf">get_num_new_matched_tokens</span><span class="p">(</span>
|
||
<span class="linenos">141</span> <span class="bp">self</span><span class="p">,</span> <span class="n">request</span><span class="p">:</span> <span class="n">LlmRequest</span><span class="p">,</span>
|
||
<span class="linenos">142</span> <span class="n">num_computed_tokens</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-></span> <span class="nb">tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]:</span>
|
||
<span class="linenos">143</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span><span class="p">[</span><span class="n">request</span><span class="o">.</span><span class="n">request_id</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
|
||
<span class="linenos">144</span>
|
||
<span class="linenos">145</span> <span class="c1"># Don't bother with sequences with partial matches.</span>
|
||
<span class="linenos">146</span> <span class="k">if</span> <span class="p">(</span><span class="n">num_computed_tokens</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
|
||
<span class="linenos">147</span> <span class="k">return</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span>
|
||
<span class="linenos">148</span>
|
||
<span class="linenos">149</span> <span class="n">computed_blocks</span> <span class="o">=</span> <span class="n">num_computed_tokens</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span>
|
||
<span class="linenos">150</span>
|
||
<span class="linenos">151</span> <span class="c1"># Get all the tokens that don't have a cache hit on device.</span>
|
||
<span class="linenos">152</span> <span class="n">remaining_tokens</span> <span class="o">=</span> <span class="n">request</span><span class="o">.</span><span class="n">get_tokens</span><span class="p">(</span><span class="mi">0</span><span class="p">)[</span><span class="n">computed_blocks</span> <span class="o">*</span>
|
||
<span class="linenos">153</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">:]</span>
|
||
<span class="linenos">154</span>
|
||
<span class="linenos">155</span> <span class="n">remaining_chunks</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_chunk_tokens</span><span class="p">(</span><span class="n">remaining_tokens</span><span class="p">)</span>
|
||
<span class="linenos">156</span>
|
||
<span class="linenos">157</span> <span class="c1"># For each chunk, check if it exists in our cache.</span>
|
||
<span class="linenos">158</span> <span class="k">for</span> <span class="n">chunk</span> <span class="ow">in</span> <span class="n">remaining_chunks</span><span class="p">:</span>
|
||
<span class="linenos">159</span> <span class="c1"># Only do full blocks.</span>
|
||
<span class="linenos">160</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">chunk</span><span class="p">)</span> <span class="o">==</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">:</span>
|
||
<span class="linenos">161</span> <span class="n">hashed_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_hash_tokens</span><span class="p">(</span><span class="n">chunk</span><span class="p">)</span>
|
||
<span class="linenos">162</span>
|
||
<span class="linenos">163</span> <span class="n">file_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_file_path</span><span class="p">(</span><span class="n">hashed_tokens</span><span class="p">)</span>
|
||
<span class="linenos">164</span>
|
||
<span class="linenos">165</span> <span class="c1"># If we get a cache hit, we want to load it into device.</span>
|
||
<span class="linenos">166</span> <span class="c1"># Otherwise, we can stop looking.</span>
|
||
<span class="linenos">167</span> <span class="k">if</span> <span class="n">file_path</span><span class="o">.</span><span class="n">exists</span><span class="p">():</span>
|
||
<span class="linenos">168</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span><span class="p">[</span><span class="n">request</span><span class="o">.</span><span class="n">request_id</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
|
||
<span class="linenos">169</span> <span class="k">else</span><span class="p">:</span>
|
||
<span class="linenos">170</span> <span class="k">break</span>
|
||
<span class="linenos">171</span>
|
||
<span class="linenos">172</span> <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
|
||
<span class="linenos">173</span> <span class="sa">f</span><span class="s2">"KV CONNECTOR: Matched </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span><span class="p">[</span><span class="n">request</span><span class="o">.</span><span class="n">request_id</span><span class="p">])</span><span class="si">}</span><span class="s2"> blocks for request </span><span class="si">{</span><span class="n">request</span><span class="o">.</span><span class="n">request_id</span><span class="si">}</span><span class="s2">"</span>
|
||
<span class="linenos">174</span> <span class="p">)</span>
|
||
<span class="linenos">175</span>
|
||
<span class="linenos">176</span> <span class="k">return</span> <span class="nb">len</span><span class="p">(</span>
|
||
<span class="linenos">177</span> <span class="bp">self</span><span class="o">.</span><span class="n">pending_loads</span><span class="p">[</span><span class="n">request</span><span class="o">.</span><span class="n">request_id</span><span class="p">])</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span> <span class="kc">False</span>
|
||
<span class="linenos">178</span>
|
||
<span class="linenos">179</span> <span class="k">def</span><span class="w"> </span><span class="nf">request_finished</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">request</span><span class="p">:</span> <span class="n">LlmRequest</span><span class="p">,</span>
|
||
<span class="linenos">180</span> <span class="n">cache_block_ids</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-></span> <span class="nb">bool</span><span class="p">:</span>
|
||
<span class="linenos">181</span> <span class="c1"># We don't do any asynchronous saving, so always return False</span>
|
||
<span class="linenos">182</span> <span class="k">return</span> <span class="kc">False</span>
|
||
<span class="linenos">183</span>
|
||
<span class="linenos">184</span> <span class="k">def</span><span class="w"> </span><span class="nf">update_state_after_alloc</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">request</span><span class="p">:</span> <span class="n">LlmRequest</span><span class="p">,</span>
|
||
<span class="linenos">185</span> <span class="n">block_ids</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">int</span><span class="p">]):</span>
|
||
<span class="linenos">186</span> <span class="k">pass</span>
|
||
<span class="linenos">187</span>
|
||
<span class="linenos">188</span>
|
||
<span class="linenos">189</span><span class="nd">@click</span><span class="o">.</span><span class="n">command</span><span class="p">()</span>
|
||
<span class="linenos">190</span><span class="nd">@click</span><span class="o">.</span><span class="n">argument</span><span class="p">(</span><span class="s2">"model"</span><span class="p">,</span> <span class="nb">type</span><span class="o">=</span><span class="nb">str</span><span class="p">)</span>
|
||
<span class="linenos">191</span><span class="k">def</span><span class="w"> </span><span class="nf">main</span><span class="p">(</span><span class="n">model</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
|
||
<span class="linenos">192</span> <span class="n">sys</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
|
||
<span class="linenos">193</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">dirname</span><span class="p">(</span><span class="vm">__file__</span><span class="p">),</span>
|
||
<span class="linenos">194</span> <span class="s2">".."</span><span class="p">,</span>
|
||
<span class="linenos">195</span> <span class="p">))</span>
|
||
<span class="linenos">196</span>
|
||
<span class="linenos">197</span> <span class="n">this_module</span> <span class="o">=</span> <span class="vm">__file__</span><span class="p">[</span><span class="vm">__file__</span><span class="o">.</span><span class="n">rfind</span><span class="p">(</span><span class="s2">"/"</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span><span class="p">:</span><span class="vm">__file__</span><span class="o">.</span><span class="n">rfind</span><span class="p">(</span><span class="s2">".py"</span><span class="p">)]</span>
|
||
<span class="linenos">198</span>
|
||
<span class="linenos">199</span> <span class="n">kv_connector_config</span> <span class="o">=</span> <span class="n">KvCacheConnectorConfig</span><span class="p">(</span>
|
||
<span class="linenos">200</span> <span class="n">connector_module</span><span class="o">=</span><span class="n">this_module</span><span class="p">,</span>
|
||
<span class="linenos">201</span> <span class="n">connector_scheduler_class</span><span class="o">=</span><span class="s2">"PersistentKvCacheConnectorLeader"</span><span class="p">,</span>
|
||
<span class="linenos">202</span> <span class="n">connector_worker_class</span><span class="o">=</span><span class="s2">"PersistentKvCacheConnectorWorker"</span><span class="p">,</span>
|
||
<span class="linenos">203</span> <span class="p">)</span>
|
||
<span class="linenos">204</span>
|
||
<span class="linenos">205</span> <span class="n">connector_cache_dir</span> <span class="o">=</span> <span class="n">TemporaryDirectory</span><span class="p">()</span>
|
||
<span class="linenos">206</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="n">CONNECTOR_CACHE_FOLDER_KEY</span><span class="p">]</span> <span class="o">=</span> <span class="n">connector_cache_dir</span><span class="o">.</span><span class="n">name</span>
|
||
<span class="linenos">207</span>
|
||
<span class="linenos">208</span> <span class="n">llm</span> <span class="o">=</span> <span class="n">LLM</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
|
||
<span class="linenos">209</span> <span class="n">backend</span><span class="o">=</span><span class="s2">"pytorch"</span><span class="p">,</span>
|
||
<span class="linenos">210</span> <span class="n">cuda_graph_config</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||
<span class="linenos">211</span> <span class="n">kv_connector_config</span><span class="o">=</span><span class="n">kv_connector_config</span><span class="p">)</span>
|
||
<span class="linenos">212</span>
|
||
<span class="linenos">213</span> <span class="n">test_text</span> <span class="o">=</span> <span class="p">(</span>
|
||
<span class="linenos">214</span> <span class="s2">"Nvidia Corporation is an American technology company headquartered in Santa Clara, California."</span>
|
||
<span class="linenos">215</span> <span class="s2">"Founded in 1993 by Jensen Huang, Chris Malachowsky, and Curtis Priem, it develops graphics processing units (GPUs), "</span>
|
||
<span class="linenos">216</span> <span class="s2">"system on a chips (SoCs), and application programming interfaces (APIs) for data science, high-performance computing, "</span>
|
||
<span class="linenos">217</span> <span class="s2">"and mobile and automotive applications. Tell me about the company."</span><span class="p">)</span>
|
||
<span class="linenos">218</span>
|
||
<span class="linenos">219</span> <span class="n">sampling_params</span> <span class="o">=</span> <span class="n">SamplingParams</span><span class="p">(</span><span class="n">max_tokens</span><span class="o">=</span><span class="mi">32</span><span class="p">)</span>
|
||
<span class="linenos">220</span>
|
||
<span class="linenos">221</span> <span class="n">output</span> <span class="o">=</span> <span class="n">llm</span><span class="o">.</span><span class="n">generate</span><span class="p">([</span><span class="n">test_text</span><span class="p">],</span> <span class="n">sampling_params</span><span class="p">)</span>
|
||
<span class="linenos">222</span> <span class="n">text0</span> <span class="o">=</span> <span class="n">output</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">text</span>
|
||
<span class="linenos">223</span>
|
||
<span class="linenos">224</span> <span class="nb">print</span><span class="p">(</span><span class="s2">"First output: "</span><span class="p">,</span> <span class="n">text0</span><span class="p">)</span>
|
||
<span class="linenos">225</span> <span class="nb">print</span><span class="p">(</span><span class="s2">"Loading new LLM instance..."</span><span class="p">)</span>
|
||
<span class="linenos">226</span>
|
||
<span class="linenos">227</span> <span class="k">del</span> <span class="n">llm</span>
|
||
<span class="linenos">228</span>
|
||
<span class="linenos">229</span> <span class="n">llm</span> <span class="o">=</span> <span class="n">LLM</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
|
||
<span class="linenos">230</span> <span class="n">backend</span><span class="o">=</span><span class="s2">"pytorch"</span><span class="p">,</span>
|
||
<span class="linenos">231</span> <span class="n">cuda_graph_config</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||
<span class="linenos">232</span> <span class="n">kv_connector_config</span><span class="o">=</span><span class="n">kv_connector_config</span><span class="p">)</span>
|
||
<span class="linenos">233</span>
|
||
<span class="linenos">234</span> <span class="n">output</span> <span class="o">=</span> <span class="n">llm</span><span class="o">.</span><span class="n">generate</span><span class="p">([</span><span class="n">test_text</span><span class="p">],</span> <span class="n">sampling_params</span><span class="p">)</span>
|
||
<span class="linenos">235</span> <span class="n">text1</span> <span class="o">=</span> <span class="n">output</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">text</span>
|
||
<span class="linenos">236</span>
|
||
<span class="linenos">237</span> <span class="nb">print</span><span class="p">(</span><span class="s2">"Second output (using connector cache): "</span><span class="p">,</span> <span class="n">text1</span><span class="p">)</span>
|
||
<span class="linenos">238</span>
|
||
<span class="linenos">239</span> <span class="k">assert</span> <span class="n">text0</span> <span class="o">==</span> <span class="n">text1</span>
|
||
<span class="linenos">240</span>
|
||
<span class="linenos">241</span> <span class="n">connector_cache_dir</span><span class="o">.</span><span class="n">cleanup</span><span class="p">()</span>
|
||
<span class="linenos">242</span>
|
||
<span class="linenos">243</span>
|
||
<span class="linenos">244</span><span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">"__main__"</span><span class="p">:</span>
|
||
<span class="linenos">245</span> <span class="n">main</span><span class="p">()</span>
|
||
</pre></div>
|
||
</div>
|
||
</section>
|
||
|
||
|
||
</article>
|
||
|
||
|
||
|
||
|
||
|
||
<footer class="prev-next-footer d-print-none">
|
||
|
||
<div class="prev-next-area">
|
||
<a class="left-prev"
|
||
href="llm_speculative_decoding.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">Speculative Decoding</p>
|
||
</div>
|
||
</a>
|
||
<a class="right-next"
|
||
href="llm_runtime.html"
|
||
title="next page">
|
||
<div class="prev-next-info">
|
||
<p class="prev-next-subtitle">next</p>
|
||
<p class="prev-next-title">Runtime Configuration Examples</p>
|
||
</div>
|
||
<i class="fa-solid fa-angle-right"></i>
|
||
</a>
|
||
</div>
|
||
</footer>
|
||
|
||
</div>
|
||
|
||
|
||
|
||
<div class="bd-sidebar-secondary"></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 August 29, 2025.</p>
|
||
|
||
<p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/15ec2b8">15ec2b8</a>.</p>
|
||
|
||
</div></div>
|
||
|
||
</div>
|
||
|
||
|
||
|
||
</div>
|
||
|
||
</footer>
|
||
</body>
|
||
</html> |