取得背景資訊快取的相關資訊

您可以瞭解脈絡快取的建立時間、最近更新時間和到期時間。如要取得與 Google Cloud 專案相關聯的所有內容快取資訊 (包括快取 ID)，請使用指令列出內容快取。如果您知道內容快取的快取 ID，可以只取得該內容快取的相關資訊。

取得脈絡快取清單

如要取得與 Google Cloud 專案相關聯的內容快取清單，您需要建立專案的區域和 Google Cloud 專案 ID。以下說明如何取得 Google Cloud 專案的內容快取清單。

Python

安裝

pip install --upgrade google-genai

詳情請參閱 SDK 參考說明文件。

設定環境變數，透過 Vertex AI 使用 Gen AI SDK：

# Replace the `GOOGLE_CLOUD_PROJECT` and `GOOGLE_CLOUD_LOCATION` values
# with appropriate values for your project.
export GOOGLE_CLOUD_PROJECT=GOOGLE_CLOUD_PROJECT
export GOOGLE_CLOUD_LOCATION=global
export GOOGLE_GENAI_USE_VERTEXAI=True

from google import genai
from google.genai.types import HttpOptions

client = genai.Client(http_options=HttpOptions(api_version="v1"))

content_cache_list = client.caches.list()

# Access individual properties of a ContentCache object(s)
for content_cache in content_cache_list:
    print(f"Cache `{content_cache.name}` for model `{content_cache.model}`")
    print(f"Last updated at: {content_cache.update_time}")
    print(f"Expires at: {content_cache.expire_time}")

# Example response:
# * Cache `projects/111111111111/locations/.../cachedContents/1111111111111111111` for
#       model `projects/111111111111/locations/.../publishers/google/models/gemini-XXX-pro-XXX`
# * Last updated at: 2025-02-13 14:46:42.620490+00:00
# * CachedContentUsageMetadata(audio_duration_seconds=None, image_count=167, text_count=153, total_token_count=43130, video_duration_seconds=None)
# ...

Go

瞭解如何安裝或更新 Go。

詳情請參閱 SDK 參考說明文件。

設定環境變數，透過 Vertex AI 使用 Gen AI SDK：

# Replace the `GOOGLE_CLOUD_PROJECT` and `GOOGLE_CLOUD_LOCATION` values
# with appropriate values for your project.
export GOOGLE_CLOUD_PROJECT=GOOGLE_CLOUD_PROJECT
export GOOGLE_CLOUD_LOCATION=global
export GOOGLE_GENAI_USE_VERTEXAI=True

import (
	"context"
	"fmt"
	"io"
	"net/http"
	"time"

	"google.golang.org/genai"
)

// listContentCache shows how to retrieve details about cached content.
func listContentCache(w io.Writer) error {
	ctx := context.Background()

	client, err := genai.NewClient(ctx, &genai.ClientConfig{
		HTTPOptions: genai.HTTPOptions{APIVersion: "v1"},
	})
	if err != nil {
		return fmt.Errorf("failed to create genai client: %w", err)
	}

	// Retrieve cached content metadata
	cache, err := client.Caches.List(ctx, &genai.ListCachedContentsConfig{
		HTTPOptions: &genai.HTTPOptions{
			Headers:    http.Header{"X-Custom-Header": []string{"example"}},
			APIVersion: "v1",
		},
	})
	if err != nil {
		return fmt.Errorf("failed to get content cache: %w", err)
	}

	// Print basic info about the cached content
	fmt.Fprintf(w, "Cache name: %s\n", cache.Name)
	fmt.Fprintf(w, "Display name: %s\n", cache.Items[0].DisplayName)
	fmt.Fprintf(w, "Model: %s\n", cache.Items[0].Model)
	fmt.Fprintf(w, "Create time: %s\n", cache.Items[0].CreateTime.Format(time.RFC3339))
	fmt.Fprintf(w, "Update time: %s\n", cache.Items[0].UpdateTime.Format(time.RFC3339))
	fmt.Fprintf(w, "Expire time: %s (in %s)\n", cache.Items[0].ExpireTime.Format(time.RFC3339), time.Until(cache.Items[0].ExpireTime).Round(time.Second))

	if cache.Items[0].UsageMetadata != nil {
		fmt.Fprintf(w, "Usage metadata: %+v\n", cache.Items[0].UsageMetadata)
	}

	// Example response:
	// Cache name: projects/111111111111/locations/us-central1/cachedContents/1234567890123456789
	// Display name: product_recommendations_prompt
	// Model: models/gemini-2.5-flash
	// Create time: 2025-04-08T02:15:23Z
	// Update time: 2025-04-08T03:05:11Z
	// Expire time: 2025-04-20T03:05:11Z (in 167h59m59s)
	// Usage metadata: &{AudioDurationSeconds:0 ImageCount:167 TextCount:153 TotalTokenCount:43124 VideoDurationSeconds:0}

	return nil
}

Java

瞭解如何安裝或更新 Java。

詳情請參閱 SDK 參考說明文件。

設定環境變數，透過 Vertex AI 使用 Gen AI SDK：

# Replace the `GOOGLE_CLOUD_PROJECT` and `GOOGLE_CLOUD_LOCATION` values
# with appropriate values for your project.
export GOOGLE_CLOUD_PROJECT=GOOGLE_CLOUD_PROJECT
export GOOGLE_CLOUD_LOCATION=global
export GOOGLE_GENAI_USE_VERTEXAI=True

import com.google.genai.Client;
import com.google.genai.types.CachedContent;
import com.google.genai.types.HttpOptions;
import com.google.genai.types.ListCachedContentsConfig;

public class ContentCacheList {

  public static void main(String[] args) {
    contentCacheList();
  }

  // Lists all cached contents
  public static void contentCacheList() {
    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests.
    try (Client client =
        Client.builder()
            .location("global")
            .vertexAI(true)
            .httpOptions(HttpOptions.builder().apiVersion("v1").build())
            .build()) {

      for (CachedContent content : client.caches.list(ListCachedContentsConfig.builder().build())) {
        content.name().ifPresent(name -> System.out.println("Name: " + name));
        content.model().ifPresent(model -> System.out.println("Model: " + model));
        content.updateTime().ifPresent(time -> System.out.println("Last updated at: " + time));
        content.expireTime().ifPresent(time -> System.out.println("Expires at: " + time));
      }
      // Example response:
      // Name: projects/111111111111/locations/global/cachedContents/1111111111111111111
      // Model:
      // projects/111111111111/locations/global/publishers/google/models/gemini-2.5-flash
      // Last updated at: 2025-07-28T21:54:19.125825Z
      // Expires at: 2025-08-04T21:54:18.328233500Z
      // ...
    }
  }
}

Node.js

安裝

npm install @google/genai

詳情請參閱 SDK 參考說明文件。

設定環境變數，透過 Vertex AI 使用 Gen AI SDK：

# Replace the `GOOGLE_CLOUD_PROJECT` and `GOOGLE_CLOUD_LOCATION` values
# with appropriate values for your project.
export GOOGLE_CLOUD_PROJECT=GOOGLE_CLOUD_PROJECT
export GOOGLE_CLOUD_LOCATION=global
export GOOGLE_GENAI_USE_VERTEXAI=True

const {GoogleGenAI} = require('@google/genai');

const GOOGLE_CLOUD_PROJECT = process.env.GOOGLE_CLOUD_PROJECT;
const GOOGLE_CLOUD_LOCATION = process.env.GOOGLE_CLOUD_LOCATION || 'global';
async function listContentCaches(
  projectId = GOOGLE_CLOUD_PROJECT,
  location = GOOGLE_CLOUD_LOCATION
) {
  const client = new GoogleGenAI({
    vertexai: true,
    project: projectId,
    location: location,
    httpOptions: {
      apiVersion: 'v1',
    },
  });

  const contentCacheList = await client.caches.list();

  // Access individual properties of a ContentCache object(s)
  const contentCacheNames = [];
  for (const contentCache of contentCacheList.pageInternal) {
    console.log(
      `Cache \`${contentCache.name}\` for model \`${contentCache.model}\``
    );
    console.log(`Last updated at: ${contentCache.updateTime}`);
    console.log(`Expires at: ${contentCache.expireTime}`);
    contentCacheNames.push(contentCache.name);
  }
  console.log(contentCacheNames);

  // Example response:
  //  * Cache `projects/111111111111/locations/us-central1/cachedContents/1111111111111111111` for
  //  model `projects/111111111111/locations/us-central1/publishers/google/models/gemini-XXX-pro-XXX`
  //  * Last updated at: 2025-02-13 14:46:42.620490+00:00
  //  * CachedContentUsageMetadata(audio_duration_seconds=None, image_count=167, text_count=153, total_token_count=43130, video_duration_seconds=None)
  // ...

  return contentCacheNames;
}

REST

以下說明如何使用 REST，將 GET 要求傳送至發布商模型端點，列出與 Google Cloud 專案相關聯的內容快取。

使用任何要求資料之前，請先替換以下項目：

• PROJECT_ID：您的專案 ID。
• LOCATION：處理建立內容快取要求的區域。

HTTP 方法和網址：

GET https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/cachedContents

如要傳送要求，請選擇以下其中一個選項：

curl

注意：下列指令假設您已執行 gcloud init 或 gcloud auth login 透過使用者帳戶登入 gcloud CLI，或是使用 Cloud Shell 自動登入 gcloud CLI。您可以執行 gcloud auth list，查看目前使用的帳戶。

執行下列指令：

curl -X GET \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     "https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/cachedContents"

PowerShell

注意：下列指令假設您已執行 gcloud init 或 gcloud auth login，透過使用者帳戶登入 gcloud CLI。您可以執行 gcloud auth list，查看目前使用的帳戶。

執行下列指令：

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method GET `
    -Headers $headers `
    -Uri "https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/cachedContents" | Select-Object -Expand Content

您應該會收到類似如下的 JSON 回應：

回應

{
  "cachedContents": [
    {
      "name": "projects/PROJECT_NUMBER/locations/us-central1/cachedContents/CACHE_ID_1",
      "model": "projects/PROJECT_ID/locations/us-central1/publishers/google/models/gemini-2.0-flash-001",
      "createTime": "2024-05-31T19:04:35.380412Z",
      "updateTime": "2024-05-31T19:04:35.380412Z",
      "expireTime": "2024-05-31T20:04:35.349680Z"
    },
    {
      "name": "projects/PROJECT_NUMBER/locations/us-central1/cachedContents/CACHE_ID_2",
      "model": "projects/PROJECT_ID/locations/us-central1/publishers/google/models/gemini-2.0-flash-001",
      "createTime": "2024-05-30T21:14:39.880235Z",
      "updateTime": "2024-05-31T00:21:15.350969Z",
      "expireTime": "2024-05-31T01:21:15.348014Z"
    },
    {
      "name": "projects/PROJECT_NUMBER/locations/us-central1/cachedContents/CACHE_ID_N",
      "model": "projects/PROJECT_ID/locations/us-central1/publishers/google/models/gemini-2.0-flash-001",
      "createTime": "2024-05-30T21:14:39.880235Z",
      "updateTime": "2024-05-31T00:21:15.350969Z",
      "expireTime": "2024-05-31T01:21:15.348014Z"
    }
  ]
}

cURL 指令範例

LOCATION="us-central1"
PROJECT_ID="PROJECT_ID"

curl \
-X GET \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
https://${LOCATION}-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/${LOCATION}/cachedContents

取得內容快取的相關資訊

如要取得單一內容快取資訊，您需要快取 ID、與內容快取相關聯的Google Cloud 專案 ID，以及處理建立內容快取要求的區域。建立脈絡快取時，系統會傳回脈絡快取的快取 ID。您也可以使用內容快取清單指令，取得與專案相關聯的每個內容快取 ID。

以下說明如何取得單一內容快取的相關資訊。

Go

在試用這個範例之前，請先按照 Vertex AI 快速入門導覽課程的操作說明設定 Go 環境。詳情請參閱 Vertex AI Go SDK for Gemini 參考說明文件。

如要向 Vertex AI 進行驗證，請設定應用程式預設憑證。詳情請參閱「 為本機開發環境設定 ADC」。

串流和非串流回應

您可以選擇模型生成串流或非串流回覆。如果是串流回覆，系統會在生成輸出權杖後立即傳送回覆。如果是非串流回覆，所有輸出權杖生成後，您就會收到所有回覆。

如要取得串流回應，請使用 GenerateContentStream 方法。

  iter := model.GenerateContentStream(ctx, genai.Text("Tell me a story about a lumberjack and his giant ox. Keep it very short."))
  

如為非串流回應，請使用 GenerateContent 方法。

  resp, err := model.GenerateContent(ctx, genai.Text("What is the average size of a swallow?"))
  

程式碼範例

import (
	"context"
	"fmt"
	"io"

	"cloud.google.com/go/vertexai/genai"
)

// getContextCache shows how to retrieve the metadata of a cached content
// contentName is the ID of the cached content to retrieve
func getContextCache(w io.Writer, contentName string, projectID, location string) error {
	// location := "us-central1"
	ctx := context.Background()

	client, err := genai.NewClient(ctx, projectID, location)
	if err != nil {
		return fmt.Errorf("unable to create client: %w", err)
	}
	defer client.Close()

	cachedContent, err := client.GetCachedContent(ctx, contentName)
	if err != nil {
		return fmt.Errorf("GetCachedContent: %w", err)
	}
	fmt.Fprintf(w, "Retrieved cached content %q", cachedContent.Name)
	return nil
}

REST

以下說明如何使用 REST，將 GET 要求傳送至發布商模型端點，列出與 Google Cloud 專案相關聯的內容快取。

使用任何要求資料之前，請先替換以下項目：

• PROJECT_ID：。
• LOCATION：處理建立脈絡快取要求的區域。
• CACHE_ID：內容快取的 ID。建立脈絡快取時，系統會傳回脈絡快取 ID。您也可以列出專案的脈絡快取，找出脈絡快取 ID。 Google Cloud 詳情請參閱「建立脈絡快取」和「列出脈絡快取」。

HTTP 方法和網址：

GET https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/cachedContents/CACHE_ID

如要傳送要求，請選擇以下其中一個選項：

curl

注意：下列指令假設您已執行 gcloud init 或 gcloud auth login 透過使用者帳戶登入 gcloud CLI，或是使用 Cloud Shell 自動登入 gcloud CLI。您可以執行 gcloud auth list，查看目前使用的帳戶。

執行下列指令：

curl -X GET \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     "https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/cachedContents/CACHE_ID"

PowerShell

注意：下列指令假設您已執行 gcloud init 或 gcloud auth login，透過使用者帳戶登入 gcloud CLI。您可以執行 gcloud auth list，查看目前使用的帳戶。

執行下列指令：

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method GET `
    -Headers $headers `
    -Uri "https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/cachedContents/CACHE_ID" | Select-Object -Expand Content

您應該會收到類似如下的 JSON 回應：

回應

{
  "name": "projects/PROJECT_NUMBER/locations/us-central1/cachedContents/CACHE_ID",
  "model": "projects/PROJECT_ID/locations/us-central1/publishers/google/models/gemini-2.0-flash-001",
  "createTime": "2024-05-31T19:04:35.380412Z",
  "updateTime": "2024-05-31T19:04:35.380412Z",
  "expireTime": "2024-05-31T20:04:35.349680Z"
}

cURL 指令範例

LOCATION="us-central1"
PROJECT_ID="PROJECT_ID"
CACHE_ID="CACHE_ID"

curl \
-X GET \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
https://${LOCATION}-aiplatform.googleapis.com/v1/projects/${PROJECT_ID}/locations/${LOCATION}/${CACHE_ID}

• 瞭解如何使用脈絡快取。
• 瞭解如何更新脈絡快取的到期時間。
• 瞭解如何刪除脈絡快取。