[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間:2025-09-04 (世界標準時間)。"],[[["\u003cp\u003eThis content provides an overview of the four APIs available with Vertex AI on Google Distributed Cloud (GDC) air-gapped: Optical Character Recognition (OCR), Speech-to-Text, Translation, and Vertex AI Workbench.\u003c/p\u003e\n"],["\u003cp\u003eYou can programmatically interact with these APIs using a service endpoint, and the Vertex AI Workbench endpoint and its discovery document can be accessed via a provided URL or \u003ccode\u003ekubectl proxy\u003c/code\u003e command.\u003c/p\u003e\n"],["\u003cp\u003eThe OCR and Translation APIs support both gRPC and REST, while Speech-to-Text supports gRPC, and Vertex AI Workbench supports Kubernetes Resource Model (KRM).\u003c/p\u003e\n"],["\u003cp\u003eGoogle offers Python client libraries for the OCR, Speech-to-Text, and Translation APIs, providing built-in features like authentication and retries, whereas REST and gRPC require custom client development.\u003c/p\u003e\n"],["\u003cp\u003eDepending on if you are using client libraries, REST, gRPC, or KRM, the type, method, and field names for the API will vary, with REST being arranged by resource hierarchies, gRPC and client libraries by services, and KRM fields using camel case.\u003c/p\u003e\n"]]],[],null,["# Vertex AI API overview\n\nThis page provides an overview of using the APIs installed with\nVertex AI on Google Distributed Cloud (GDC) air-gapped and its reference\ndocumentation.\n\nService endpoint and discovery document\n---------------------------------------\n\nA service endpoint is required to interact programmatically with the\nVertex AI APIs.\n\n### Get the Vertex AI Workbench endpoint and discovery document\n\nThe API endpoint for the Vertex AI Workbench KRM API is:\n\n`https://`\u003cvar translate=\"no\"\u003eENDPOINT\u003c/var\u003e`/apis/aiplatform.gdc.goog/v1`\n\nReplace \u003cvar translate=\"no\"\u003eENDPOINT\u003c/var\u003e with the API endpoint of the\nManagement API server.\n\nTo access the Vertex AI Workbench discovery document, perform one of\nthe following actions:\n\n- Enter the endpoint URL in a browser.\n\n- Run the `kubectl proxy` command in a tool such as curl to open a proxy to the\n Management API server on your local machine. After that command is running,\n enter the following URL in your browser:\n\n `http://127.0.0.1:8001/apis/aiplatform.gdc.goog/v1`.\n\n### Get the pre-trained APIs endpoints\n\nTo get the endpoints for the pre-trained APIs,\n[view service status and endpoints](/distributed-cloud/hosted/docs/latest/gdch/application/ao-user/vertex-ai-api-status).\n\nREST, gRPC, KRM, and client libraries\n-------------------------------------\n\nYou can access the pre-trained APIs using gRPC or one of the provided client\nlibraries. The client libraries are built on gRPC.\n\nAlternatively, you access some pre-trained APIs using REST.\n\nYou can manage Vertex AI Workbench using its Kubernetes Resource\nModel (KRM) API.\n\n### Client libraries\n\nVertex AI provides Python client libraries for pre-trained APIs. The\nfollowing table shows a comparison of advantages and disadvantages of using\nclient libraries:\n\n### REST\n\nThe OCR and Translation APIs support REST. For more\ninformation, see the REST API references for these services:\n\n- [OCR REST API reference](/distributed-cloud/hosted/docs/latest/gdch/apis/vertex-ai/ocr/rest/overview)\n\n- [Translation REST API reference](/distributed-cloud/hosted/docs/latest/gdch/apis/vertex-ai/translation/rest)\n\n| **Note:** The Speech-to-Text and Vertex AI Workbench APIs don't support REST.\n\nThe following table shows a comparison of advantages and disadvantages of using\nREST:\n\n### gRPC\n\nPre-trained APIs support gRPC. For more information about the generic\ndescriptions of the types, methods, and fields generated for a gRPC library, see\nthe following gRPC reference:\n\n- [OCR gRPC reference](/distributed-cloud/hosted/docs/latest/gdch/apis/vertex-ai/ocr/rpc)\n- [Speech-to-Text gRPC reference](/distributed-cloud/hosted/docs/latest/gdch/apis/vertex-ai/speech-to-text/rpc)\n- [Translation gRPC reference](/distributed-cloud/hosted/docs/latest/gdch/apis/vertex-ai/translation/rpc)\n\n| **Note:** The Vertex AI Workbench API doesn't support gRPC.\n\nThe following table shows a comparison of advantages and disadvantages of using\ngRPC:\n\n### KRM\n\nThe Vertex AI Workbench API supports KRM. For more information, see\nthe [Vertex AI Workbench KRM API reference](/distributed-cloud/hosted/docs/latest/gdch/apis/vertex-ai/workbench/krm/vertex-ai-workbench-krm-ref).\n\nType, method, and field names\n-----------------------------\n\nDepending on whether you are using client libraries, REST, gRPC, or KRM, the\ntype, method, and field names for the API vary in the following ways:\n\n- REST is arranged by resource hierarchies and their methods.\n- Client libraries and gRPC are arranged by services and their methods.\n- KRM field names use camelCase, but the API service accepts either camelCase or snake_case.\n- REST and gRPC field names use snake_case.\n- Client library field names use either title case, camelCase, or snake_case, depending on which name is idiomatic for the language.\n\nREST and protocol buffers\n-------------------------\n\nWhen calling the REST API, the default value behavior for protocol\nbuffers might result in missing fields in a JSON response. These fields are\nset to the default value, so they are not included in the response."]]