Functionality related to Google Cloud Platform and AI Studio

​
Chat models

​
Gemini Models

Access Gemini models such as gemini-1.5-pro and gemini-2.0-flex through the ChatGoogleGenerativeAI, or if using VertexAI, via the ChatVertexAI class.
See this section for general instructions on installing integration packages.
npm
npm install @langchain/google-genai @langchain/core
Configure your API key.
export GOOGLE_API_KEY=your-api-key

import { ChatGoogleGenerativeAI } from "@langchain/google-genai";

const model = new ChatGoogleGenerativeAI({
  model: "gemini-pro",
  maxOutputTokens: 2048,
});

// Batch and stream are also supported
const res = await model.invoke([
  [
    "human",
    "What would be a good company name for a company that makes colorful socks?",
  ],
]);
More recent Gemini models support image inputs:
const visionModel = new ChatGoogleGenerativeAI({
  model: "gemini-2.0-flash",
  maxOutputTokens: 2048,
});
const image = fs.readFileSync("./hotdog.jpg").toString("base64");
const input2 = [
  new HumanMessage({
    content: [
      {
        type: "text",
        text: "Describe the following image.",
      },
      {
        type: "image_url",
        image_url: `data:image/png;base64,${image}`,
      },
    ],
  }),
];

const res = await visionModel.invoke(input2);
Click here for the @langchain/google-genai specific integration docs
The value of image_url must be a base64 encoded image (e.g., data:image/png;base64,abcd124).

​
Gemma

Access the gemma-3-27b-it model through AI Studio using the ChatGoogle class. (This class is a superclass of the ChatVertexAI class that works with both Vertex AI and the AI Studio APIs.)
Since Gemma is an open model, it may also be available from other platforms including Ollama.
npm
npm install @langchain/google-gauth @langchain/core
Configure your API key. 
export GOOGLE_API_KEY=your-api-key

import { ChatGoogle } from "@langchain/google-gauth";

const model = new ChatGoogle({
  model: "gemma-3-27b-it",
});

const res = await model.invoke([
  {
    role: "user",
    content:
      "What would be a good company name for a company that makes colorful socks?",
  },
]);

​
Third Party Models

See above for setting up authentication through Vertex AI to use these models. Anthropic Claude models are also available through the Vertex AI platform. See here for more information about enabling access to the models and the model names to use. PaLM models are no longer supported.

​
Vector Store

Vertex AI Vector Search, formerly known as Vertex AI Matching Engine, provides the industry’s leading high-scale low latency vector database. These vector databases are commonly referred to as vector similarity-matching or an approximate nearest neighbor (ANN) service. 
import { MatchingEngine } from "@langchain/community/vectorstores/googlevertexai";

​
Postgres Vector Store

The PostgresVectorStore module from the @langchain/google-cloud-sql-pg package provides a way to use the CloudSQL for PostgresSQL to store vector embeddings using the class. 
$ yarn add @langchain/google-cloud-sql-pg
Set your environment variables: 
PROJECT_ID="your-project-id"
REGION="your-project-region"
INSTANCE_NAME="your-instance"
DB_NAME="your-database-name"
DB_USER="your-database-user"
PASSWORD="your-database-password"
Create a DB connection through the PostgresEngine class: 
const engine: PostgresEngine = await PostgresEngine.fromInstance(
  process.env.PROJECT_ID ?? "",
  process.env.REGION ?? "",
  process.env.INSTANCE_NAME ?? "",
  process.env.DB_NAME ?? "",
  peArgs
);
Initialize the vector store table: 
await engine.initVectorstoreTable(
  "my_vector_store_table",
  768,
  vectorStoreArgs
);
Create a vector store instance: 
const vectorStore = await PostgresVectorStore.initialize(
  engine,
  embeddingService,
  "my_vector_store_table",
  pvectorArgs
);

​
Tools

  • Set up a Custom Search Engine, following these instructions
  • Get an API Key and Custom Search Engine ID from the previous step, and set them as environment variables GOOGLE_API_KEY and GOOGLE_CSE_ID respectively
There exists a GoogleCustomSearch utility which wraps this API. To import this utility: 
import { GoogleCustomSearch } from "@langchain/community/tools/google_custom_search";
We can easily load this wrapper as a Tool (to use with an Agent). We can do this with: 
const tools = [new GoogleCustomSearch({})];
// Pass this variable into your agent.

​
Chat history

​
Postgres Chat Message History

The PostgresChatMessageHistory from the @langchain/google-cloud-sql-pg package provides a way to use the CloudSQL for PostgresSQL to store messages and provide conversation history. 
$ yarn add @langchain/google-cloud-sql-pg
Note: See the [Postgres Vector Store](#Postgres Vector Store) section on this page to learn how to install the package and initialize a DB connection. Initialize the chat history table: 
await engine.initChatHistoryTable("chat_message_table");
Create a chat message history instance: 
const historyInstance = await PostgresChatMessageHistory.initialize(
  engine,
  "test",
  "chat_message_table"
);

​
Document Loaders

​
Postgres Loader

The PostgresLoader from @langchain/google-cloud-sql-pg provides a way to use the CloudSQL for PostgresSQL to load data as LangChain Documents. Note: See the [Postgres Vector Store](#Postgres Vector Store) section on this page to learn how to install the package and initialize a DB connection. Create a loader instance: 
const documentLoaderArgs: PostgresLoaderOptions = {
  tableName: "test_table_custom",
  contentColumns: ["fruit_name", "variety"],
  metadataColumns: [
    "fruit_id",
    "quantity_in_stock",
    "price_per_unit",
    "organic",
  ],
  format: "text",
};

const documentLoaderInstance = await PostgresLoader.initialize(
  PEInstance,
  documentLoaderArgs
);

const documents = await documentLoaderInstance.load();