プロビジョンド スループットは、Vertex AI でサポートされている生成 AI モデルのスループットを予約する、固定料金の固定期間サブスクリプションです。このサブスクリプションは、複数の期間で利用できます。スループットを予約するには、モデルと、モデルが実行される利用可能なロケーションを指定する必要があります。
[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["わかりにくい","hardToUnderstand","thumb-down"],["情報またはサンプルコードが不正確","incorrectInformationOrSampleCode","thumb-down"],["必要な情報 / サンプルがない","missingTheInformationSamplesINeed","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["その他","otherDown","thumb-down"]],["最終更新日 2025-09-04 UTC。"],[],[],null,["# Provisioned Throughput overview\n\nThis page explains what Provisioned Throughput is and when to use Provisioned Throughput.\n\nIntroduction to Provisioned Throughput\n--------------------------------------\n\nProvisioned Throughput is a fixed-cost, fixed-term subscription\navailable in several term-lengths that reserves throughput for\n[supported generative AI models](/vertex-ai/generative-ai/docs/supported-models) on Vertex AI.\nTo reserve your throughput, you must specify the model and [available\nlocations](/vertex-ai/generative-ai/docs/learn/locations#available-regions) in which the model\nruns.\n\nWhen to use Provisioned Throughput\n----------------------------------\n\nIf any of the following considerations apply to your use case, consider using\nProvisioned Throughput:\n\n- You are building real-time generative AI production applications, such as chatbots and agents.\n- Your critical workloads consistently require high throughput. Throughput measurement depends on the model.\n- You want to provide a consistent and predictable experience for users of your applications.\n- You want deterministic generative AI costs by paying a fixed monthly or weekly price with control of overages.\n\nProvisioned Throughput is one of two ways to consume your\ngenerative AI models. The second way is pay-as-you-go, which is also referred to\nas [on-demand](/vertex-ai/generative-ai/docs/error-code-429#troubleshoot-dynamic-shared-quota).\n\nWhat's next\n-----------\n\n- [Supported models](/vertex-ai/generative-ai/docs/supported-models) using Provisioned Throughput."]]