Google 集成

使用 LangChain Python 与 Google 集成。

本页面涵盖了 LangChain 与Google Gemini, Google Cloud, and other Google products (such as Google Maps, YouTube, and more).

Unified SDK & package consolidation

As of langchain-google-genai 4.0.0, this package uses the consolidated google-genai SDK and now supports both the Gemini Developer API and Vertex AI backends.

The langchain-google-vertexai package remains supported for Vertex AI platform-specific features (Model Garden, Vector Search, evaluation services, etc.).

Read the full announcement and migration guide.

Not sure which package to use?

Google Generative AI (Gemini API & Vertex AI)

Access Google Gemini models via the Gemini Developer API or Vertex AI. The backend is selected automatically based on your configuration.

• Gemini Developer API: Quick setup with API key, ideal for individual developers and rapid prototyping
• Vertex AI: Enterprise features with Google Cloud integration (requires GCP project)

Use the langchain-google-genai package for chat models, LLMs, and embeddings.

See integrations.

Google Cloud (Vertex AI Platform Services)

Access Vertex AI platform-specific services beyond Gemini models: Model Garden (Llama, Mistral, Anthropic), evaluation services, and specialized vision models.

Use the langchain-google-vertexai package for platform services and specific packages (e.g., langchain-google-community, langchain-google-cloud-sql-pg) for other cloud services like databases and storage.

See integrations.

See Google’s guide on migrating from the Gemini API to Vertex AI for more details on the differences.

---

Google Generative AI

Access Google Gemini models via the Gemini Developer API or Vertex AI 使用 unified langchain-google-genai 包。

聊天模型

ChatGoogleGenerativeAI

Google Gemini chat models via Gemini Developer API or Vertex AI.

开始使用

大语言模型 (LLM)

GoogleGenerativeAI

Gemini models 使用 (legacy) LLM text completion interface.

开始使用

向量嵌入模型

GoogleGenerativeAIEmbeddings

Gemini embedding models via Gemini Developer API or Vertex AI.

开始使用

---

Google Cloud

Access Vertex AI platform-specific services including Model Garden (Llama, Mistral, Anthropic), Vector Search, evaluation services, and specialized vision models.

For Gemini models, use ChatGoogleGenerativeAI from langchain-google-genai. The classes below focus on Vertex AI platform services not available in the consolidated SDK.

聊天模型

ChatAnthropicVertex

Anthropic on Vertex AI Model Garden

开始使用

ChatVertexAI (deprecated)

Deprecated—Use ChatGoogleGenerativeAI for Gemini models instead.

from langchain_google_vertexai import ChatVertexAI

VertexModelGardenLlama

Llama on Vertex AI Model Garden

from langchain_google_vertexai.model_garden_maas.llama import VertexModelGardenLlama

VertexModelGardenMistral

Mistral on Vertex AI Model Garden

from langchain_google_vertexai.model_garden_maas.mistral import VertexModelGardenMistral

GemmaChatLocalHF

Local Gemma model loaded from HuggingFace.

from langchain_google_vertexai.gemma import GemmaChatLocalHF

GemmaChatLocalKaggle

Local Gemma model loaded from Kaggle.

from langchain_google_vertexai.gemma import GemmaChatLocalKaggle

GemmaChatVertexAIModelGarden

Gemma on Vertex AI Model Garden

from langchain_google_vertexai.gemma import GemmaChatVertexAIModelGarden

VertexAIImageCaptioningChat

Image captioning model as a chat interface.

from langchain_google_vertexai.vision_models import VertexAIImageCaptioningChat

VertexAIImageEditorChat

Edit images given a prompt. Currently supports mask-free editing only.

from langchain_google_vertexai.vision_models import VertexAIImageEditorChat

VertexAIImageGeneratorChat

Generate images from a prompt.

from langchain_google_vertexai.vision_models import VertexAIImageGeneratorChat

VertexAIVisualQnAChat

Visual question answering model as a chat interface.

from langchain_google_vertexai.vision_models import VertexAIVisualQnAChat

大语言模型 (LLM)

(Legacy) string-in, string-out LLM interface.

VertexAIModelGarden

Hundreds of OSS models via Vertex AI Model Garden.

开始使用

VertexAI (deprecated)

Deprecated—Use GoogleGenerativeAI for Gemini models instead.

from langchain_google_vertexai import VertexAI

Gemma local from Hugging Face

Local Gemma model loaded from HuggingFace.

from langchain_google_vertexai.gemma import GemmaLocalHF

Gemma local from Kaggle

Local Gemma model loaded from Kaggle.

from langchain_google_vertexai.gemma import GemmaLocalKaggle

Gemma on Vertex AI Model Garden

from langchain_google_vertexai.gemma import GemmaVertexAIModelGarden

Vertex AI image captioning

Image captioning model as an LLM interface.

from langchain_google_vertexai.vision_models import VertexAIImageCaptioning

向量嵌入模型

VertexAIEmbeddings (deprecated)

Deprecated—Use GoogleGenerativeAIEmbeddings instead.

from langchain_google_vertexai import VertexAIEmbeddings

文档加载器

AlloyDB for PostgreSQL

PostgreSQL-compatible database on Google Cloud.

开始使用

BigQuery

Serverless data warehouse.

开始使用

Bigtable

Key-value and wide-column store for structured and semi-structured data.

开始使用

Cloud SQL for MySQL

Managed MySQL database.

开始使用

Cloud SQL for SQL Server

Managed SQL Server database.

开始使用

Cloud SQL for PostgreSQL

Managed PostgreSQL database.

开始使用

Cloud Storage (directory)

Load documents from a GCS bucket directory.

开始使用

Cloud Storage (file)

Load a single document from GCS.

开始使用

El Carro for Oracle Workloads

Oracle databases on Kubernetes via El Carro.

开始使用

Firestore (Native Mode)

NoSQL document database.

开始使用

Firestore (Datastore Mode)

Firestore in Datastore mode.

开始使用

Memorystore for Redis

Managed Redis service.

开始使用

Spanner

Globally distributed relational database.

开始使用

Speech-to-Text

Transcribe audio files.

开始使用

Cloud Vision loader

Load data using Google Cloud Vision API.

from langchain_google_community.vision import CloudVisionLoader

文档转换器

Document AI

Extract structured data from unstructured documents.

开始使用

Google Translate

Translate text and HTML via Cloud Translation API.

开始使用

向量存储

Store and search vectors using Google Cloud databases and Vertex AI Vector Search.

AlloyDB for PostgreSQL

PostgreSQL-compatible vector store on AlloyDB.

开始使用

BigQuery Vector Search

Semantic search using GoogleSQL with vector indexes.

开始使用

Memorystore for Redis

Vector store on Memorystore for Redis.

开始使用

Spanner

Vector store on Cloud Spanner.

开始使用

Bigtable

Vector store on Cloud Bigtable.

开始使用

Firestore (Native Mode)

Vector store on Firestore.

开始使用

Cloud SQL for MySQL

Vector store on Cloud SQL for MySQL.

开始使用

Cloud SQL for PostgreSQL

Vector store on Cloud SQL for PostgreSQL.

开始使用

Vertex AI Vector Search

Formerly known as Vertex AI Matching Engine, provides a low latency vector database. These vector databases are commonly referred to as vector similarity-matching or an approximate nearest neighbor (ANN) service.

开始使用

Vertex AI Vector Search + Datastore

Vector search with Datastore for document storage.

开始使用

检索器

Vertex AI Search

Generative AI powered search via Vertex AI Search.

开始使用

Document AI Warehouse

Search, store, and manage documents using Document AI Warehouse.

开始使用

Other retrievers

from langchain_google_community import VertexAIMultiTurnSearchRetriever
from langchain_google_community import VertexAISearchRetriever
from langchain_google_community import VertexAISearchSummaryTool

工具

Integrate agents with various Google Cloud services.

Text-to-Speech

Synthesize natural-sounding speech with 100+ voices.

开始使用

回调

Track LLM/Chat model usage.

Vertex AI callback handler

Track VertexAI usage info.

from langchain_google_vertexai.callbacks import VertexAICallbackHandler

Google BigQuery

See the documentation for more details.

from langchain_google_community.callbacks.bigquery_callback import BigQueryCallbackHandler

Evaluators

Evaluate model outputs using Vertex AI.

VertexPairWiseStringEvaluator

Pair-wise evaluation using Vertex AI models.

from langchain_google_vertexai.evaluators.evaluation import VertexPairWiseStringEvaluator

VertexStringEvaluator

Single prediction evaluation using Vertex AI models.

from langchain_google_vertexai.evaluators.evaluation import VertexStringEvaluator

---

Other Google products

Integrations with various Google services beyond the core Cloud Platform.

文档加载器

Google Drive

Load files from Google Drive. Currently supports Google Docs.

开始使用

向量存储

ScaNN (Local Index)

Efficient local vector similarity search at scale.

开始使用

检索器

Google Drive

Retrieve documents from Google Drive.

开始使用

工具

Google Search

Web search via Google Custom Search Engine (CSE).

开始使用

Google Drive

Interact with Google Drive.

开始使用

Google Finance

Query financial data.

开始使用

Google Jobs

Query job listings.

开始使用

Google Lens

Visual searches.

开始使用

Google Places

Search for places.

开始使用

Google Scholar

Search academic papers.

开始使用

Google Trends

Query Google Trends data.

开始使用

MCP

MCP Toolbox

Connect to databases including Cloud SQL and AlloyDB.

开始使用

工具包

Gmail

Create, search, and send emails via the Gmail API.

开始使用

Chat loaders

Gmail

Load chat history from Gmail threads.

开始使用

---

3rd party integrations

Access Google services via unofficial third-party APIs.

Search

SearchApi

API access to Google search results, YouTube, and more.

开始使用

SerpApi

API access to Google search results.

开始使用

Serper.dev

API access to Google search results.

开始使用

cloro

Google Search results with AI Overview support.

开始使用

YouTube

Search tool

Search YouTube videos without the official API.

开始使用

Audio loader

Download audio from YouTube videos.

开始使用

Transcripts loader

Load video transcripts.

开始使用

---

将这些文档连接 到 Claude、VSCode 等工具，通过 MCP 获取实时答案。

在 GitHub 上编辑此页面或提交 issue.