不過,隨著生成式 AI 解決方案日趨成熟,您可能需要一個平台,以便建構及部署端對端生成式 AI 應用程式和解決方案。Google Cloud 提供全方位的工具生態系統,協助開發人員充分運用生成式 AI 的強大功能,從應用程式開發的初始階段,到應用程式部署、應用程式代管,以及大規模管理複雜資料,都能得心應手。
Google Cloud 的 Vertex AI 平台提供全套機器學習運作工具,可簡化 AI 模型的運用、部署和監控程序,提升效率和可靠性。此外,與資料庫、DevOps 工具、記錄、監控和 IAM 的整合,可提供全面性的方法,管理整個生成式 AI 生命週期。
# To install the Python SDK, use this CLI command:# pip install google-generativeaiimportgoogle.generativeaiasgenaifromgoogle.generativeaiimportGenerativeModelAPI_KEY="API_KEY"genai.configure(api_key=API_KEY)
# To install the Python SDK, use this CLI command:# pip install google-genaifromgoogleimportgenaiPROJECT_ID="PROJECT_ID"LOCATION="LOCATION"# e.g. us-central1client=genai.Client(project=PROJECT_ID,location=LOCATION,vertexai=True)
根據文字提示生成文字的程式碼範例
Gemini Developer API
Vertex AI 的 Gemini API
model=GenerativeModel("gemini-2.0-flash")response=model.generate_content("The opposite of hot is")print(response.text)# The opposite of hot is cold.
fromgoogleimportgenaifromgoogle.genai.typesimportHttpOptionsclient=genai.Client(http_options=HttpOptions(api_version="v1"))response=client.models.generate_content(model="gemini-2.5-flash",contents="How does AI work?",)print(response.text)# Example response:# Okay, let's break down how AI works. It's a broad field, so I'll focus on the ...## Here's a simplified overview:# ...
使用文字和圖片生成文字的程式碼範例
Gemini Developer API
Vertex AI 的 Gemini API
importPIL.Imagemultimodal_model=GenerativeModel("gemini-2.0-flash")image=PIL.Image.open("image.jpg")response=multimodal_model.generate_content(["What is this picture?",image])print(response.text)# A cat is shown in this picture.
fromgoogleimportgenaifromgoogle.genai.typesimportHttpOptions,Partclient=genai.Client(http_options=HttpOptions(api_version="v1"))response=client.models.generate_content(model="gemini-2.5-flash",contents=["What is shown in this image?",Part.from_uri(file_uri="gs://cloud-samples-data/generative-ai/image/scones.jpg",mime_type="image/jpeg",),],)print(response.text)# Example response:# The image shows a flat lay of blueberry scones arranged on parchment paper. There are ...
產生多輪對話的程式碼範例
Gemini Developer API
Vertex AI 的 Gemini API
model=GenerativeModel("gemini-2.0-flash")chat=model.start_chat()print(chat.send_message("How are you?").text)print(chat.send_message("What can you do?").text)
fromgoogleimportgenaifromgoogle.genai.typesimportHttpOptions,ModelContent,Part,UserContentclient=genai.Client(http_options=HttpOptions(api_version="v1"))chat_session=client.chats.create(model="gemini-2.5-flash",history=[UserContent(parts=[Part(text="Hello")]),ModelContent(parts=[Part(text="Great to meet you. What would you like to know?")],),],)response=chat_session.send_message("Tell me a story.")print(response.text)# Example response:# Okay, here's a story for you:# ...
將提示遷移至 Vertex AI Studio
Google AI Studio 提示資料會儲存在 Google 雲端硬碟資料夾中。本節說明如何將提示遷移至 Vertex AI Studio。
[[["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)。"],[],[],null,["# Migrate from the Gemini Developer API to the Gemini API in Vertex AI\n\nIf you are new to Gemini, using the [quickstarts](https://ai.google.dev/gemini-api/docs/quickstart?lang=python)\nis the fastest way to get started.\n\n\nHowever, as your generative AI solutions mature, you may need a platform for building and\ndeploying generative AI applications and solutions end to end. Google Cloud provides a\ncomprehensive ecosystem of tools to enable developers to harness the power of generative AI,\nfrom the initial stages of app development to app deployment, app hosting, and managing complex\ndata at scale.\n\n\nGoogle Cloud's Vertex AI platform offers a suite of MLOps tools that streamline usage, deployment,\nand monitoring of AI models for efficiency and reliability. Additionally, integrations with\ndatabases, DevOps tools, logging, monitoring, and IAM provide a holistic approach to managing the\nentire generative AI lifecycle. \n\n#### Common use cases for Google Cloud offerings\n\n\nHere are some examples of common use cases that are well-suited for Google Cloud offerings.\n\n- Productionize your apps and solutions. Products like [Cloud Run functions](https://cloud.google.com/functions/docs/concepts/overview) and [Cloud Run](https://cloud.google.com/run/docs/overview/what-is-cloud-run) lets you to deploy apps with enterprise-grade scale, security and privacy. Find more details about security and privacy on the [Security, Privacy, and Cloud Compliance on Google Cloud](https://cloud.google.com/security) guide.\n- Use Vertex AI for end to end MLOps capabilities from tuning to vector similarity-search and ML pipelines.\n- Trigger your LLM call with event-driven architecture with [Cloud Run functions](https://cloud.google.com/functions/docs/concepts/overview) or [Cloud Run](https://cloud.google.com/run/docs/overview/what-is-cloud-run).\n- Monitor usage of your app with [Cloud Logging](https://cloud.google.com/logging/docs) and [BigQuery](https://cloud.google.com/logging/docs/export/bigquery).\n- Store your data with enterprise-grade security, at scale with services like [BigQuery](https://cloud.google.com/bigquery/docs), [Cloud Storage](https://cloud.google.com/storage/docs/introduction), and [Cloud SQL](https://cloud.google.com/sql).\n- Perform retrieval-augmented generation (RAG) using data in the cloud with [BigQuery](https://cloud.google.com/bigquery/docs) or [Cloud Storage](https://cloud.google.com/storage/docs/introduction).\n- Create and schedule data pipelines. You can [schedule jobs](https://cloud.google.com/scheduler/docs/schedule-run-cron-job) using [Cloud Scheduler](https://cloud.google.com/scheduler/docs/overview).\n- Apply LLMs to your data in the cloud. If you store data in Cloud Storage or BigQuery, you can prompt LLMs over that data. For example to extract information, summarize or ask questions about it.\n- Leverage Google Cloud [data governance/residency](https://cloud.google.com/learn/what-is-data-governance) policies to manage your data lifecycle.\n\nDifferences between the Gemini Developer API and the Gemini API in Vertex AI\n----------------------------------------------------------------------------\n\n\nThe following table summarizes the main differences between the\nGemini Developer API and the Vertex AI Gemini API to help you\ndecide which option is right for your use case:\n\nMigrate to Gemini API in Vertex AI\n----------------------------------\n\nThis section shows how to migrate from the Gemini Developer API to\nthe Gemini API in Vertex AI.\n\n**Considerations when migrating**\n\nConsider the following when migrating:\n\n- You can use your existing Google Cloud project (the same one you used to\n generate your Gemini API key) or you can create a new\n [Google Cloud project](/resource-manager/docs/creating-managing-projects).\n\n- Supported regions might differ between the Gemini Developer API and\n the Gemini API in Vertex AI. See the list of\n [supported regions for generative AI on Google Cloud](/vertex-ai/generative-ai/docs/learn/locations).\n\n- Any models you created in Google AI Studio need to be retrained in\n Vertex AI.\n\n### Start using Vertex AI Studio\n\nThe process you follow to migrate to Gemini API in Vertex AI is different, depending\non if you already have a Google Cloud account or you are new to Google Cloud.\n| **Note:** Google AI Studio and the Gemini Developer API are available only in [specific regions and languages](https://ai.google.dev/available_regions). If you aren't located in a supported region, you can't start using the Gemini API in Vertex AI.\n\nTo learn how migrate to the Gemini API in Vertex AI, click one of the following\ntabs, depending on your Google Cloud account status: \n\n### Already use Google Cloud\n\n1. Sign in to [Google AI Studio](https://aistudio.google.com/app/waitlist/97445851).\n2. At the bottom of the left navigation pane, click **Build with Vertex AI on Google Cloud**.\n\n The **Try Vertex AI and Google Cloud for free** page opens.\n3. Click **Agree \\& Continue**.\n\n The **Get Started with Vertex AI studio** dialog appears.\n4. To enable the APIs required to run Vertex AI, click **Agree \\&\n Continue**.\n\n The Vertex AI console appears. To learn how to migrate your data\n from Google AI studio, see [Migrate Prompts](#migrate-prompts).\n\n### New to Google Cloud\n\n1. Sign in to [Google AI Studio](https://aistudio.google.com/app/waitlist/97445851).\n2. At the bottom of the left navigation pane, click **Build with Vertex AI on Google Cloud**.\n\n The **Create an account to get started with Google Cloud** page opens.\n3. Click **Agree \\& Continue**.\n\n The **Let's confirm your identity** page appears.\n4. Click **Start Free**.\n\n The **Get Started with Vertex AI studio** dialog appears.\n5. To enable the APIs required to run Vertex AI, click **Agree \\&\n Continue**.\n\n6. Optional: To learn how to migrate your data from Google AI studio, see Migrate\n Prompts on this page [Migrate Prompts](#migrate-prompts).\n\n### Python: Migrate to the Gemini API in Vertex AI\n\n\nThe following sections show code snippets to help you migrate your Python code to use the\nGemini API in Vertex AI.\n\n#### Vertex AI Python SDK Setup\n\n\nOn Vertex AI, you don't need an API key. Instead, Gemini on Vertex AI is managed using IAM access,\nwhich controls permission for a user, a group, or a service account to call the Gemini API\nthrough the Vertex AI SDK.\n\n\nWhile there are [many ways\nto authenticate](https://cloud.google.com/docs/authentication#auth-decision-tree), the easiest method for authenticating in a development environment is to\n[install the Google Cloud CLI](https://cloud.google.com/sdk/docs/install)\nthen use your user credentials to\n[sign in to the CLI](https://cloud.google.com/docs/authentication/gcloud#local).\n\n\nTo make inference calls to Vertex AI, you must also make sure that your user or service account has\nthe [Vertex AI\nUser role](https://cloud.google.com/vertex-ai/docs/general/access-control#aiplatform.user).\n\n#### Code example to install the client\n\n#### Code example to generate text from text prompt\n\n#### Code example to generate text from text and image\n\n#### Code example to generate multi-turn chat\n\n### Migrate prompts to Vertex AI Studio\n\nYour Google AI Studio prompt data is saved in a Google Drive folder. This\nsection shows how to migrate your prompts to Vertex AI Studio.\n\n1. Open [Google Drive](https://drive.google.com).\n2. Navigate to the **AI_Studio** folder where the prompts are stored.\n3. Download your prompts from Google Drive to a local directory.\n\n | **Note:** Prompts downloaded from Google Drive are in the text (`txt`) format. Before you upload them to Vertex AI Studio, convert them to JSON files. To do this, change the file extension from `.txt` to `.json`.\n4. Open [Vertex AI Studio](https://console.cloud.google.com/vertex-ai/generative) in the Google Cloud console.\n\n5. In the **Vertex AI** menu, click **Prompt management**.\n\n6. Click **Import prompt**.\n\n7. In the **Prompt file** field, click **Browse** and select a prompt from\n your local directory.\n\n To upload prompts in bulk, you must manually combine your prompts into a\n single JSON file.\n8. Click **Upload**.\n\n The prompts are uploaded to the **My Prompts** tab.\n\n### Upload training data to Vertex AI Studio\n\nTo migrate your training data to Vertex AI, you need to upload your data to a Cloud Storage bucket. For more information, see\n[Introduction to tuning](https://cloud.google.com/vertex-ai/generative-ai/docs/models/tune-models).\n\nDelete unused API Keys\n----------------------\n\nIf you no longer need to use your Gemini API key for the\nGemini Developer API, then follow security best practices and delete\nit.\n\nTo delete an API key:\n\n1. Open the\n [Google Cloud API Credentials](https://console.cloud.google.com/apis/credentials)\n page.\n\n2. Find the API key that you want to delete and click the **Actions** icon.\n\n3. Select **Delete API key**.\n\n4. In the **Delete credential** modal, select **Delete**.\n\n Deleting an API key takes a few minutes to propagate. After\n propagation completes, any traffic using the deleted API key is rejected.\n\n| **Important:** If you delete a key that's still used in production and need to recover it, see [`gcloud beta services api-keys undelete`](/sdk/gcloud/reference/beta/services/api-keys/undelete).\n\nWhat's next\n-----------\n\n- Try a quickstart tutorial using [Vertex AI Studio](/vertex-ai/generative-ai/docs/start/quickstarts/quickstart) or the [Vertex AI API](/vertex-ai/generative-ai/docs/start/quickstarts/quickstart-multimodal)."]]