[[["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)：2024-11-26。"],[],[],null,["# Deployments and endpoints\n\nGoogle and Partner models and generative AI features on Vertex AI are\nexposed as specific [regional endpoints](/about/locations) and a global\nendpoint. Global endpoints cover the entire world and provide higher\navailability and reliability than single regions.\n\nNote that model endpoints don't guarantee region availability or in-region ML\nprocessing. For information about data residency, see\n[Data residency](/vertex-ai/generative-ai/docs/learn/data-residency).\n\nGlobal endpoint\n---------------\n\nSelecting a global endpoint for your requests can improve overall availability\nwhile reducing resource exhausted (429) errors. Don't use the global endpoint if\nyou have ML processing requirements, because you can't control or know which\nregion your ML processing requests are sent to when a request is made.\n\n### Supported models\n\nUsage of the global endpoint is supported for the following Google models:\n\n- [Gemini 2.5 Flash Image Preview](/vertex-ai/generative-ai/docs/models/gemini/2-5-flash#image)\n- [Gemini 2.5 Flash-Lite](/vertex-ai/generative-ai/docs/models/gemini/2-5-flash-lite)\n- [Gemini 2.5 Pro](/vertex-ai/generative-ai/docs/models/gemini/2-5-pro)\n- [Gemini 2.5 Flash](/vertex-ai/generative-ai/docs/models/gemini/2-5-flash)\n- [Gemini 2.0 Flash](/vertex-ai/generative-ai/docs/models/gemini/2-0-flash)\n- [Gemini 2.0 Flash-Lite](/vertex-ai/generative-ai/docs/models/gemini/2-0-flash-lite)\n\nFor more information about partner models, see [Google Cloud partner model\nendpoint locations](#genai-partner-models)\n\n### Use the global endpoint\n\nTo use the global endpoint, exclude the location from the endpoint name and\nconfigure the location of the resource to `global`. For example, the following\nis global endpoint URL: \n\n https://aiplatform.googleapis.com/v1/projects/test-project/locations/global/publishers/google/models/gemini-2.0-flash-001:generateContent\n\nFor the [Google Gen AI SDK](/vertex-ai/generative-ai/docs/sdks/overview), create a client that uses the `global` location: \n\n client = genai.Client(\n vertexai=True, project='\u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e', location='global'\n )\n\nFor the [Vertex AI SDK for Python](/vertex-ai/generative-ai/docs/reference/libraries),\ninitialize the SDK using the `global` location: \n\n import https://cloud.google.com/python/docs/reference/vertexai/latest/\n from vertexai.generative_models import https://cloud.google.com/python/docs/reference/vertexai/latest/vertexai.preview.generative_models.GenerativeModel.html\n\n https://cloud.google.com/python/docs/reference/vertexai/latest/.init(project='\u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e', location='global')\n\n### Limitations\n\nThe following capabilities are not available when using the global endpoint:\n\n- Tuning\n- Batch prediction\n- Retrieval-augmented generation (RAG) corpus (RAG requests are supported)\n\nUsage of the global endpoint with Provisioned Throughput is\navailable only for the following models:\n\nGoogle model endpoint locations\n-------------------------------\n\nGoogle model endpoints for Generative AI on Vertex AI are available\nin the following regions.\n**Important:** Starting April 29, 2025, Gemini 1.5 Pro and Gemini 1.5 Flash are not available in projects that have no prior usage of these models, including new projects. For details, see [Model versions and lifecycle](/vertex-ai/generative-ai/docs/learn/model-versions). \n\n### United States\n\n### Canada\n\n### South America\n\n### Europe\n\n### Asia Pacific\n\n### Middle East\n\n### Global\n\n\\* Region is available only while using [Single Zone\nProvisioned Throughput](/vertex-ai/generative-ai/docs/provisioned-throughput/szpt)\n\n+ Supervised fine tuning isn't supported in this region.\n\nGoogle Cloud partner model endpoint locations\n---------------------------------------------\n\nGoogle serves requests from the region that you specified. For some models,\nGoogle also offers a global endpoint to improve overall availability and reduce\nerror rates. The global endpoint can have a separate set of quotas from the\nregional endpoint and doesn't support data residency requirements. For more\ninformation, see the \"Regional and global endpoint\" section in [Vertex AI partner models for\nMaaS](/vertex-ai/generative-ai/docs/partner-models/use-partner-models).\n\nPartner model endpoints for Generative AI on Vertex AI are available in\nthe following regions: \n\n### United States\n\n### Europe\n\n### Asia Pacific\n\n### Global\n\nWhat's next\n-----------\n\n- For a notebook tutorial that demonstrates the global endpoint, see [Intro to\n the Vertex AI global\n endpoint](https://github.com/GoogleCloudPlatform/generative-ai/blob/main/gemini/global-endpoint/intro_global_endpoint.ipynb).\n- Learn more about Generative AI on Vertex AI [data\n residency](/vertex-ai/generative-ai/docs/learn/data-residency).\n- Learn about [Google Cloud regions](/docs/geography-and-regions).\n- Learn more about [security controls by\n feature](/vertex-ai/generative-ai/docs/genai-security-controls).\n- Learn about the models that provide Generative AI on Vertex AI support. See [Generative AI foundational model\n reference](/vertex-ai/generative-ai/docs/model-reference/overview).\n- Learn about [Vertex AI locations](/vertex-ai/docs/general/locations)."]]
