Documentation: https://googleapis.github.io/js-genai/
The Google Gen AI JavaScript SDK is designed for TypeScript and JavaScript developers to build applications powered by Gemini. The SDK supports both the Gemini Developer API and Vertex AI.
The Google Gen AI SDK is designed to work with Gemini 2.0 features.
Note
SDK Preview: See: Preview Launch.
Caution
API Key Security: Avoid exposing API keys in client-side code. Use server-side implementations in production environments.
- Node.js version 18 or later
To install the SDK, run the following command:
npm install @google/genaiThe simplest way to get started is to using an API key from Google AI Studio:
import {GoogleGenAI} from '@google/genai';
const GEMINI_API_KEY = process.env.GEMINI_API_KEY;
const ai = new GoogleGenAI({apiKey: GEMINI_API_KEY});
async function main() {
const response = await ai.models.generateContent({
model: 'gemini-2.0-flash-001',
contents: 'Why is the sky blue?',
});
console.log(response.text);
}
main();The package contents are also available unzipped in the
package/ directory of the bucket, so an equivalent web example is:
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Using My Package</title>
</head>
<body>
<script type="module">
import {GoogleGenAI} from 'https://cdn.jsdelivr.net/npm/@google/genai@latest/+esm'
const ai = new GoogleGenAI({apiKey:"GEMINI_API_KEY"});
async function main() {
const response = await ai.models.generateContent({
model: 'gemini-2.0-flash-001',
contents: 'Why is the sky blue?',
});
console.log(response.text);
}
main();
</script>
</body>
</html>The Google Gen AI SDK provides support for both the Google AI Studio and Vertex AI implementations of the Gemini API.
For server-side applications, initialize using an API key, which can be acquired from Google AI Studio:
import { GoogleGenAI } from '@google/genai';
const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'});Caution
API Key Security: Avoid exposing API keys in client-side code. Use server-side implementations in production environments.
In the browser the initialization code is identical:
import { GoogleGenAI } from '@google/genai';
const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'});Sample code for VertexAI initialization:
import { GoogleGenAI } from '@google/genai';
const ai = new GoogleGenAI({
vertexai: true,
project: 'your_project',
location: 'your_location',
});All API features are accessed through an instance of the GoogleGenAI classes.
The submodules bundle together related API methods:
ai.models: Usemodelsto query models (generateContent,generateImages, ...), or examine their metadata.ai.caches: Create and managecachesto reduce costs when repeatedly using the same large prompt prefix.ai.chats: Create local statefulchatobjects to simplify multi turn interactions.ai.files: Uploadfilesto the API and reference them in your prompts. This reduces bandwidth if you use a file many times, and handles files too large to fit inline with your prompt.ai.live: Start alivesession for real time interaction, allows text + audio + video input, and text or audio output.
More samples can be found in the github samples directory.
For quicker, more responsive API interactions use the generateContentStream
method which yields chunks as they're generated:
import {GoogleGenAI} from '@google/genai';
const GEMINI_API_KEY = process.env.GEMINI_API_KEY;
const ai = new GoogleGenAI({apiKey: GEMINI_API_KEY});
async function main() {
const response = await ai.models.generateContentStream({
model: 'gemini-2.0-flash-001',
contents: 'Write a 100-word poem.',
});
for await (const chunk of response) {
console.log(chunk.text);
}
}
main();To let Gemini to interact with external systems, you can provide provide
functionDeclaration objects as tools. To use these tools it's a 4 step
- Declare the function name, description, and parameters
- Call
generateContentwith function calling enabled - Use the returned
FunctionCallparameters to call your actual function - Send the result back to the model (with history, easier in
ai.chat) as aFunctionResponse
import {GoogleGenAI, FunctionCallingConfigMode, FunctionDeclaration, Type} from '@google/genai';
const GEMINI_API_KEY = process.env.GEMINI_API_KEY;
async function main() {
const controlLightDeclaration: FunctionDeclaration = {
name: 'controlLight',
parameters: {
type: Type.OBJECT,
description: 'Set the brightness and color temperature of a room light.',
properties: {
brightness: {
type: Type.NUMBER,
description:
'Light level from 0 to 100. Zero is off and 100 is full brightness.',
},
colorTemperature: {
type: Type.STRING,
description:
'Color temperature of the light fixture which can be `daylight`, `cool`, or `warm`.',
},
},
required: ['brightness', 'colorTemperature'],
},
};
const ai = new GoogleGenAI({apiKey: GEMINI_API_KEY});
const response = await ai.models.generateContent({
model: 'gemini-2.0-flash-001',
contents: 'Dim the lights so the room feels cozy and warm.',
config: {
toolConfig: {
functionCallingConfig: {
// Force it to call any function
mode: FunctionCallingConfigMode.ANY,
allowedFunctionNames: ['controlLight'],
}
},
tools: [{functionDeclarations: [controlLightDeclaration]}]
}
});
console.log(response.functionCalls);
}
main();The SDK is curently in a preview launch stage, per Google's launch stages this means:
At Preview, products or features are ready for testing by customers. Preview offerings are often publicly announced, but are not necessarily feature-complete, and no SLAs or technical support commitments are provided for these. Unless stated otherwise by Google, Preview offerings are intended for use in test environments only. The average Preview stage lasts about six months.