Vertex AI Studio は、組み込みのエクスペリエンスと専用のコラボレーション ワークスペース環境を通じて、モデル、設定、システム指示の検出と改良、プロンプトの改良を効率的に行うことができます。
このページでは、Vertex AI Studio に実装されている主な機能について説明します。これらの機能により、革新的なアイデアを生成 AI ソリューションに変換できます。たとえば、カスタマー サービスとサポート、コンテンツ作成、ヘルスケア、金融サービス、ソフトウェア開発に役立つソリューションなどです。
ロール
次のロールを割り当てることで、ユーザーが Vertex AI Studio を介してアクセスできる Google Cloud サービスを制御できます。
役割
説明
管理者
技術用語を使用してエージェントとやり取りし、Vertex AI を Google Cloud エコシステムに接続する設定を担当します。 Google Cloud
Builder
デベロッパー ツールを使用する ML デベロッパーとアプリ デベロッパーが含まれます。エージェントは、Vertex AI に限定されたスコープでデベロッパー固有の言語を使用します。
[[["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 2025-09-10 UTC。"],[],[],null,["Vertex AI Studio offers an efficient experience for\ndiscovering and refining models, settings, system instructions, and prompt\nrefinement through built-in experiences and a dedicated collaborative workspace\nenvironment.\n\nThis page outlines the key features implemented within\nVertex AI Studio that facilitate transforming innovative ideas\ninto generative AI solutions, for example, solutions that benefit customer\nservice and support, content creation, healthcare, financial services, and\nsoftware development.\n\nRoles\n\nYou can control which Google Cloud services that users can access through\nVertex AI Studio by assigning the following roles:\n\n| **Role** | **Description** |\n|---------------|-------------------------------------------------------------------------------------------------------------------------------------------------------|\n| Administrator | Responsible for Google Cloud setup that interacts with agents using technical language to connect Vertex AI with the Google Cloud ecosystem. |\n| Builder | Includes the ML developers and App developers, who use the developer tools. Agents use developer-specific language with a scope limited to Vertex AI. |\n\nFeatures and experiences\n\nThis section describes Vertex AI Studio capabilities.\n\nOnboarding and administrative\n\nThe onboarding and administrative features let you access and set up your\nenvironment.\n\n| **Feature** | **Description** |\n|--------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| Access to API key | You can access your API key quickly. |\n| Left navigation redesign | The left menu includes a Floating Action Button (FAB) that lets you open different playgrounds, view recent sessions, and access other options. |\n| Getting started menu | A menu for new users with steps to properly activate their accounts, which include trying a prompt, adding a collaborator, and getting an API key. |\n| New experience opt-in | A dialog that prompts existing users to try the latest signin and can revert to their earlier UI in settings. |\n| Settings menu | A settings menu that includes options to switch between the previous and latest UI, change the appearance to use the light or dark system, generate and access API keys, and manage billing. |\n\nStudio discovery and development\n\nThe studio discovery and development features enhance the developer workflow.\n\n| **Feature** | **Description** |\n|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| Interactive canvas view | A canvas that illustrates the artifact to be retrieved for development, which includes a preview and a code view. |\n| Minimap | A summarized history of prompt and response headers that can be turned off. |\n| Slash commands | Keyboard shortcuts for developers, which include the following commands: - **`/ask`**: You can get Google Cloud and general help. - **`/clear`**: You can clear the current conversation. - **`/compare`**: You can see how changes impact a response. - **`/model`**: You can manage your model selection. - **`/prompt`**: You can manage and modify prompts. |\n| Side-by-side comparison | An updated comparison feature for models, system instructions, and prompts, which lets you compare multiple prompts and your responses simultaneously. |\n| Upload large files | The ability to upload local or Cloud Storage files up to 20 MB. |\n| Generate code files | A feature that generates application code files, which are then accessible in the canvas pane. |\n| Download code files | The ability to download generated code files as a zip file. |\n| Undo or Redo | Undo or redo functionality is available for actions like model settings and prompt changes. |\n| In-App help tool | An in-app help tool accessible through the `/ask` command or the prompt field. |\n| Prompt gallery | An updated prompt gallery with app samples. |\n\nBuilt-in tools and capabilities\n\nThis section lists the built-in tools and capabilities within\nVertex AI Studio that automate a significant portion of\nthe work for developers. Vertex AI Studio integrates with Google Cloud, which\nprovides an effortless experience that's collaborative and agentic.\n\n| **Feature** | **Description** |\n|-----------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| Natural language refinement | For prompt refinement, you can describe how you want to change your prompt using natural language, and the system automatically optimizes the prompt for you. |\n| Optimize system instruction | You can use a one-click feature that asks Gemini to automatically optimize a system instruction based on the prompt. |\n| Refine prompt response | You can use the `/prompt` command to provide feedback and to create a well-formatted revised prompt. |\n| Help-me-write tool | You can use the `/prompt` command and a described intent to create a well-formatted prompt, system instruction, and to recommend a model. |\n\nNext steps\n\n- To learn more about agents, see [AI Applications](/vertex-ai/generative-ai/docs/agent-builder/overview).\n- See an [Introduction to\n prompting](/vertex-ai/generative-ai/docs/learn/prompts/introduction-prompt-design).\n- Learn how to [Design multimodal\n prompts](/vertex-ai/generative-ai/docs/multimodal/design-multimodal-prompts).\n- Learn how to [Design chat prompts](/vertex-ai/generative-ai/docs/chat/chat-prompts)."]]
