[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-09-04。"],[],[],null,["# Overview of custom training options in Vertex AI\n\nCompare Vertex AI custom training and Ray on Vertex AI\n------------------------------------------------------\n\nVertex AI offers two options for custom training, Vertex AI\ncustom training and Ray on Vertex AI. This page provides context for\nhelping choose between these two options. \n\nKey differences between Vertex AI custom training and Ray on Vertex AI\n----------------------------------------------------------------------\n\nVertex AI custom training is a broader service managing various\ntraining methods, while Ray on Vertex AI specifically uses the\nRay distributed computing framework. \n\nSummary\n-------\n\nIf you need to use the power of distributed computing with the Ray\nframework within the Google Cloud environment, Ray on Vertex AI is\nthe service to use. Ray on Vertex AI can be considered a specific\ntool within the larger Vertex AI ecosystem, particularly useful\nfor highly scalable and distributed workloads.\n\nIf you need a more general-purpose managed platform for various model\ntraining approaches, including automated options, custom code execution, and\nhyperparameter tuning, the broader Vertex AI custom training services\nare useful."]]