進行初步測試時,您可以將 API 金鑰寫死在程式碼中,但這只是暫時做法,因為並不安全。如需以硬式編碼方式提供 API 金鑰的範例,請參閱「明確提供 API 金鑰」一節。
將 API 金鑰設為環境變數
如果您設定環境變數 GEMINI_API_KEY 或 GOOGLE_API_KEY,使用 Gemini API 程式庫時,用戶端會自動擷取 API 金鑰。建議您只設定其中一個變數,但如果兩個都設定,GOOGLE_API_KEY 會優先採用。
如果您使用 REST API 或瀏覽器上的 JavaScript,則必須明確提供 API 金鑰。
以下說明如何在不同作業系統中,將 API 金鑰在本機設為環境變數 GEMINI_API_KEY。
Linux/macOS - Bash
Bash 是常見的 Linux 和 macOS 終端機設定。如要檢查是否有設定檔,請執行下列指令:
~/.bashrc
如果回應為「No such file or directory」,您需要建立這個檔案,並執行下列指令來開啟檔案,或使用 zsh:
touch~/.bashrcopen~/.bashrc
接著,您需要新增下列匯出指令,設定 API 金鑰:
exportGEMINI_API_KEY=<YOUR_API_KEY_HERE>
儲存檔案後,請執行下列指令來套用變更:
source~/.bashrc
macOS - Zsh
Zsh 是常見的 Linux 和 macOS 終端機設定。如要檢查是否有設定檔,請執行下列指令:
~/.zshrc
如果回應為「No such file or directory」,您需要建立這個檔案,並執行下列指令來開啟檔案,或使用 bash:
touch~/.zshrcopen~/.zshrc
接著,您需要新增下列匯出指令,設定 API 金鑰:
exportGEMINI_API_KEY=<YOUR_API_KEY_HERE>
儲存檔案後,請執行下列指令來套用變更:
source~/.zshrc
Windows
在系統設定中搜尋「環境變數」
編輯「使用者變數」(適用於目前使用者) 或「系統變數」(適用於所有使用者,請謹慎使用)。
建立變數並新增 export GEMINI_API_KEY=your_key_here
套用變更
明確提供 API 金鑰
在某些情況下,您可能需要明確提供 API 金鑰。例如:
您要進行簡單的 API 呼叫,且偏好對 API 金鑰進行硬式編碼。
您希望明確控管,而不必依賴 Gemini API 程式庫自動探索環境變數
您使用的環境不支援環境變數 (例如網頁),或是您正在發出 REST 呼叫。
以下範例說明如何明確提供 API 金鑰:
Python
fromgoogleimportgenaiclient=genai.Client(api_key="YOUR_API_KEY")response=client.models.generate_content(model="gemini-2.5-flash",contents="Explain how AI works in a few words")print(response.text)
JavaScript
import{GoogleGenAI}from"@google/genai";constai=newGoogleGenAI({apiKey:"YOUR_API_KEY"});asyncfunctionmain(){constresponse=awaitai.models.generateContent({model:"gemini-2.5-flash",contents:"Explain how AI works in a few words",});console.log(response.text);}main();
Go
packagemainimport("context""fmt""log""google.golang.org/genai")funcmain(){ctx:=context.Background()client,err:=genai.NewClient(ctx,&genai.ClientConfig{APIKey:"YOUR_API_KEY",Backend:genai.BackendGeminiAPI,})iferr!=nil{log.Fatal(err)}result,err:=client.Models.GenerateContent(ctx,"gemini-2.5-flash",genai.Text("Explain how AI works in a few words"),nil,)iferr!=nil{log.Fatal(err)}fmt.Println(result.Text())}
Java
packagecom.example;importcom.google.genai.Client;importcom.google.genai.types.GenerateContentResponse;publicclassGenerateTextFromTextInput{publicstaticvoidmain(String[]args){Clientclient=Client.builder().apiKey("YOUR_API_KEY").build();GenerateContentResponseresponse=client.models.generateContent("gemini-2.5-flash","Explain how AI works in a few words",null);System.out.println(response.text());}}
[[["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:2025-09-08 (世界標準時間)。"],[],[],null,["# Using Gemini API keys\n\nTo use the Gemini API, you need an API key. You can create a key for free with a\nfew clicks in [Google AI Studio](https://aistudio.google.com/app/apikey).\n\nOnce you have an API key, you have the following options to connect to the\nGemini API:\n\n- [Setting your API key as an environment variable](#set-api-env-var)\n- [Providing your API key explicitly](#provide-api-key-explicitly)\n\nFor initial testing, you can hard code an API key, but this should only be\ntemporary since it's not secure. You can find examples for hard coding the API\nkey in [Providing API key explicitly](#provide-api-key-explicitly) section.\n\nSetting API key as environment variable\n---------------------------------------\n\nIf you set the environment variable `GEMINI_API_KEY` or `GOOGLE_API_KEY`, the\nAPI key will automatically be picked up by the client when using one of the\n[Gemini API libraries](/gemini-api/docs/libraries). It's recommended that you\nset only one of those variables, but if both are set, `GOOGLE_API_KEY` takes\nprecedence.\n\nIf you're using the REST API, or JavaScript on the browser, you will need to\nprovide the API key explicitly.\n\nHere is how you can set your API key locally as the environment variable\n`GEMINI_API_KEY` with different operating systems. \n\n### Linux/macOS - Bash\n\nBash is a common Linux and macOS terminal configuration. You can check if\nyou have a configuration file for it by running the following command: \n\n ~/.bashrc\n\nIf the response is \"No such file or directory\", you will need to create this\nfile and open it by running the following commands, or use `zsh`: \n\n touch ~/.bashrc\n open ~/.bashrc\n\nNext, you need to set your API key by adding the following export command: \n\n export GEMINI_API_KEY=\u003cYOUR_API_KEY_HERE\u003e\n\nAfter saving the file, apply the changes by running: \n\n source ~/.bashrc\n\n### macOS - Zsh\n\nZsh is a common Linux and macOS terminal configuration. You can check if\nyou have a configuration file for it by running the following command: \n\n ~/.zshrc\n\nIf the response is \"No such file or directory\", you will need to create this\nfile and open it by running the following commands, or use `bash`: \n\n touch ~/.zshrc\n open ~/.zshrc\n\nNext, you need to set your API key by adding the following export command: \n\n export GEMINI_API_KEY=\u003cYOUR_API_KEY_HERE\u003e\n\nAfter saving the file, apply the changes by running: \n\n source ~/.zshrc\n\n### Windows\n\n1. Search for \"Environment Variables\" in the system settings\n2. Edit either \"User variables\" (for current user) or \"System variables\" (for all users - use with caution).\n3. Create the variable and add `export GEMINI_API_KEY=your_key_here`\n4. Apply the changes\n\nProviding API key explicitly\n----------------------------\n\nIn some cases, you may want to explicitly provide an API key. For example:\n\n- You're doing a simple API call and prefer hard coding the API key.\n- You want explicit control without having to rely on automatic discovery of environment variables by the Gemini API libraries\n- You're using an environment where environment variables are not supported (e.g web) or you are making REST calls.\n\nBelow are examples for how you can provide an API key explicitly: \n\n### Python\n\n from google import genai\n\n client = genai.Client(api_key=\"\u003cvar translate=\"no\"\u003eYOUR_API_KEY\u003c/var\u003e\")\n\n response = client.models.generate_content(\n model=\"gemini-2.5-flash\", contents=\"Explain how AI works in a few words\"\n )\n print(response.text)\n\n### JavaScript\n\n import { GoogleGenAI } from \"@google/genai\";\n\n const ai = new GoogleGenAI({ apiKey: \"\u003cvar translate=\"no\"\u003eYOUR_API_KEY\u003c/var\u003e\" });\n\n async function main() {\n const response = await ai.models.generateContent({\n model: \"gemini-2.5-flash\",\n contents: \"Explain how AI works in a few words\",\n });\n console.log(response.text);\n }\n\n main();\n\n### Go\n\n package main\n\n import (\n \"context\"\n \"fmt\"\n \"log\"\n \"google.golang.org/genai\"\n )\n\n func main() {\n ctx := context.Background()\n client, err := genai.NewClient(ctx, &genai.ClientConfig{\n APIKey: \"\u003cvar translate=\"no\"\u003eYOUR_API_KEY\u003c/var\u003e\",\n Backend: genai.BackendGeminiAPI,\n })\n if err != nil {\n log.Fatal(err)\n }\n\n result, err := client.Models.GenerateContent(\n ctx,\n \"gemini-2.5-flash\",\n genai.Text(\"Explain how AI works in a few words\"),\n nil,\n )\n if err != nil {\n log.Fatal(err)\n }\n fmt.Println(result.Text())\n }\n\n### Java\n\n package com.example;\n\n import com.google.genai.Client;\n import com.google.genai.types.GenerateContentResponse;\n\n public class GenerateTextFromTextInput {\n public static void main(String[] args) {\n Client client = Client.builder().apiKey(\"\u003cvar translate=\"no\"\u003eYOUR_API_KEY\u003c/var\u003e\").build();\n\n GenerateContentResponse response =\n client.models.generateContent(\n \"gemini-2.5-flash\",\n \"Explain how AI works in a few words\",\n null);\n\n System.out.println(response.text());\n }\n }\n\n### REST\n\n curl \"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash:generateContent?key=$\u003cvar translate=\"no\"\u003eYOUR_API_KEY\u003c/var\u003e\" \\\n -H 'Content-Type: application/json' \\\n -X POST \\\n -d '{\n \"contents\": [\n {\n \"parts\": [\n {\n \"text\": \"Explain how AI works in a few words\"\n }\n ]\n }\n ]\n }'\n\nKeep your API key secure\n------------------------\n\nTreat your Gemini API key like a password. If compromised, others can use your\nproject's quota, incur charges (if billing is enabled), and access your\nprivate data, such as files.\n\n### Critical security rules\n\n- **Never commit API keys to source control.** Do not check your API key into version control systems like Git.\n\n- **Never expose API keys on the client-side.** Do not use your API key directly\n in web or mobile apps in production. Keys in client-side code\n (including our JavaScript/TypeScript libraries and REST calls) can be\n extracted.\n\n### Best practices\n\n- **Use server-side calls with API keys** The most secure way to use your API\n key is to call the Gemini API from a server-side application where the key\n can be kept confidential.\n\n- **Use ephemeral tokens for client-side access (Live API only):** For direct\n client-side access to the Live API, you can use ephemeral tokens. They come with\n lower security risks and can be suitable for production use. Review\n [ephemeral tokens](/gemini-api/docs/ephemeral-tokens) guide for more information.\n\n- **Consider adding restrictions to your key:** You can limit a key's permissions\n by adding [API key restrictions](https://cloud.google.com/api-keys/docs/add-restrictions-api-keys#add-api-restrictions).\n This minimizes the potential damage if the key is ever leaked.\n\nFor some general best practices, you can also review this\n[support article](https://support.google.com/googleapi/answer/6310037)."]]