請注意，在此情境中，「每位使用者的要求數」是指來自單一 IP 位址的要求。如果預期多位使用者會透過 NAT 同時使用應用程式，您可能需要考慮這點。
4. 使用 Firebase ML API 金鑰呼叫 Cloud API
最後，在應用程式中設定 Firebase ML，使用新的 API 金鑰。
由於 Firebase ML API 金鑰允許未經驗證的 Cloud Vision API 存取權，因此請務必將金鑰保密，以免遭到未經授權的使用，導致帳單帳戶產生費用。為此，您應避免在應用程式二進位檔中加入 API 金鑰。請改為在應用程式執行階段驗證已登入的使用者是否為已知良好使用者，然後再從伺服器擷取 API 金鑰。
即使遵守這些做法，API 金鑰仍有可能遭盜用。您應採取相關措施，盡量減輕金鑰遭盜用造成的影響，例如如上所述減少每個使用者的 API 配額、實作金鑰輪替政策，以及為不同使用者群組核發不同的金鑰。
應用程式安全取得 API 金鑰後，如要呼叫 Firebase ML Cloud API，請指定金鑰：
Swift
ifletcloudVisionKey=getYourApiKey(){// See note above about securing your API keyletoptions=VisionCloudDetectorOptions()options.apiKeyOverride=cloudVisionKeyletcloudDetector=Vision.vision().cloudLandmarkDetector(options:options)}
Objective-C
NSString*cloudVisionKey=[selfgetYourApiKey];// See note above about securing your API keyif(cloudVisionKey!=nil){FIRVisionCloudDetectorOptions*options=[[FIRVisionCloudDetectorOptionsalloc]init];options.APIKeyOverride=cloudVisionKey;FIRVisionCloudLandmarkDetector*landmarkDetector=[visioncloudLandmarkDetectorWithOptions:options];}
[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["缺少我需要的資訊","missingTheInformationINeed","thumb-down"],["過於複雜/步驟過多","tooComplicatedTooManySteps","thumb-down"],["過時","outOfDate","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["示例/程式碼問題","samplesCodeIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間：2025-07-25 (世界標準時間)。"],[],[],null,["\u003cbr /\u003e\n\nIf your Apple app uses one of Firebase ML's cloud APIs, before you launch your app\nin production, you should take some additional steps to prevent unauthorized\nAPI access.\n\n1. Reduce the scope of existing API keys\n\nFirst, configure your existing API keys to disallow access to the Cloud Vision\nAPI:\n\n1. Open the [Credentials](https://console.cloud.google.com/apis/credentials?project=_) page of the\n Google Cloud console. When prompted, select your project.\n\n2. For each existing API key in the list, open the editing view.\n\n3. In the *API restrictions* section, select **Restrict key** , then add to the\n list all of the APIs to which you want the API key to have access. Make sure\n to ***not*** include the Cloud Vision API.\n\n When you configure an API key's *API restrictions* , you are explicitly\n declaring the APIs to which the key has access. **By default, when the *API\n restrictions* section has *Don't restrict key* selected, an API key can be\n used to access any API that is enabled for the project.**\n\nNow, your existing API keys will not grant access to cloud ML services, but each\nkey will continue to work for any APIs that you added to its *API restrictions*\nlist.\n\nNote that if you enable any additional APIs in the future, you must add them to\nthe *API restrictions* list for the applicable API key.\n\n2. Create a new API key for use with Firebase ML\n\nNext, create a new API key for Firebase ML that only allows calls to the\nCloud Vision API:\n\n1. Return to the [Credentials](https://console.cloud.google.com/apis/credentials) page. Be sure your\n Firebase project is still selected.\n\n2. Click **Create credentials \\\u003e API key** . Take note of the new API key, then\n click **Restrict key**.\n\n3. In the *API restrictions* section, select **Restrict key** , then add to the\n list ***only*** the Cloud Vision API.\n\nThis API key grants access only to the Cloud Vision API and can be used by\nFirebase ML to access cloud-based models.\n\n3. Recommended: Reduce the Cloud Vision API's quota\n\nTo help mitigate the consequences of a compromised key, you should reduce the\nCloud Vision API's per-user quota from its default setting. To do so:\n\n1. Open the [Cloud Vision API Quotas](https://console.cloud.google.com/apis/api/vision.googleapis.com/quotas?project=_) page of the\n Google Cloud console. When prompted, select your project.\n\n2. In the **Requests** section, set the **Requests per minute per user** quota\n to something reasonable for your app. For example, if your app involves\n uploading a picture of a document to get back its text, it's unlikely that a\n user will do that more than once every few seconds, so a quota of 30-40\n would probably be safe.\n\n Note that in this context \"requests per user\" refers to requests from a\n single IP address. You might need to consider this if you expect multiple\n users to use your app at the same time from behind a NAT.\n\n4. Call Cloud APIs using your Firebase ML API key\n\nFinally, in your app, configure Firebase ML to use your new API key.\n\nBecause the Firebase ML API key allows unauthenticated access to the Cloud Vision\nAPI, it's important to keep the key confidential to prevent unauthorized use and\ncharges to your billing account. To do so, you should refrain from including\nyour API key in your app binary. Instead, at app runtime, verify that a known\ngood user is signed in, and only then, retrieve the API key from a server.\n\nEven when these practices are observed, it is possible for an API key to be\ncompromised. You should take steps to help mitigate the consequences of a\ncompromised key, such as reducing the API's per-user quota as described above,\nimplementing key rotation policies, and issuing different keys to different\ngroups of users.\n\nAfter your app has safely acquired the API key, when you want to call an\nFirebase ML Cloud API, specify the key: \n\nSwift \n\n if let cloudVisionKey = getYourApiKey() { // See note above about securing your API key\n let options = VisionCloudDetectorOptions()\n options.apiKeyOverride = cloudVisionKey\n let cloudDetector = Vision.vision().cloudLandmarkDetector(options: options)\n }\n\nObjective-C \n\n NSString *cloudVisionKey = [self getYourApiKey]; // See note above about securing your API key\n if (cloudVisionKey != nil) {\n FIRVisionCloudDetectorOptions *options =\n [[FIRVisionCloudDetectorOptions alloc] init];\n options.APIKeyOverride = cloudVisionKey;\n FIRVisionCloudLandmarkDetector *landmarkDetector =\n [vision cloudLandmarkDetectorWithOptions:options];\n }\n\nIn addition, you should follow the general advice in\n[Securing an API key](https://cloud.google.com/docs/authentication/api-keys#securing_an_api_key).\n\nNext steps\n\nSee the [launch checklist](/support/guides/launch-checklist) for information on\npreparing your app to launch when using other Firebase features."]]
