ChatVertexAI integration | LangChain 简体中文

Danger

Deprecated

This integration is deprecated and will be removed in a future release. Please use ChatGoogleGenerativeAI instead. See the full release notes and migration guide.

Gemini Enterprise Agent Platform exposes all foundational models available in Google Cloud, like gemini-3.1-pro-preview, gemini-3.7-flash, etc. For a full and updated list of available models visit Gemini Enterprise Agent Platform documentation.

Info

Gemini Enterprise Agent Platform vs. Gemini API

The Gemini Enterprise Agent Platform integration is separate from the Google Gemini API. This page showcases an enterprise version of Gemini through Google Cloud Platform (GCP).

Tip

API Reference

For detailed documentation of all features and configuration options, head to the ChatVertexAI API reference.

Overview

Integration details

Class	Package	Serializable	JS support	Downloads	Version
`ChatVertexAI`	`langchain-google-vertexai`	beta	✅

Model features

Tool calling	Structured output	Image input	Audio input	Video input	Token-level streaming	Native async	Token usage	Logprobs
✅	✅	✅	✅	✅	✅	✅	✅	✅

Setup

To access Gemini Enterprise Agent Platform models you'll need to create a Google Cloud Platform account, set up credentials, and install the langchain-google-vertexai integration package.

Credentials

To use the integration you must either:

Have credentials configured for your environment (gcloud, workload identity, etc...)
Store the path to a service account JSON file as the GOOGLE_APPLICATION_CREDENTIALS environment variable

This codebase uses the google.auth library which first looks for the application credentials variable mentioned above, and then looks for system-level auth.

For more information, see the google.auth API reference.

To enable automated tracing of your model calls, set your LangSmith API key:

os.environ["LANGSMITH_API_KEY"] = getpass.getpass("Enter your LangSmith API key: ")
os.environ["LANGSMITH_TRACING"] = "true"

Installation

The LangChain Gemini Enterprise Agent Platform integration lives in the langchain-google-vertexai package:

pip install -qU langchain-google-vertexai

Instantiation

Now we can instantiate our model object and generate chat completions:

from langchain_google_vertexai import ChatVertexAI

llm = ChatVertexAI(
    model="gemini-3.7-flash",
    temperature=0,
    max_tokens=None,
    max_retries=6,
    stop=None,
    # other params...
)

Invocation

messages = [
    (
        "system",
        "You are a helpful assistant that translates English to French. Translate the user sentence.",
    ),
    ("human", "I love programming."),
]
ai_msg = llm.invoke(messages)
ai_msg

AIMessage(content="J'adore programmer. \n", response_metadata={'is_blocked': False, 'safety_ratings': [{'category': 'HARM_CATEGORY_HATE_SPEECH', 'probability_label': 'NEGLIGIBLE', 'blocked': False}, {'category': 'HARM_CATEGORY_DANGEROUS_CONTENT', 'probability_label': 'NEGLIGIBLE', 'blocked': False}, {'category': 'HARM_CATEGORY_HARASSMENT', 'probability_label': 'NEGLIGIBLE', 'blocked': False}, {'category': 'HARM_CATEGORY_SEXUALLY_EXPLICIT', 'probability_label': 'NEGLIGIBLE', 'blocked': False}], 'usage_metadata': {'prompt_token_count': 20, 'candidates_token_count': 7, 'total_token_count': 27}}, id='run-7032733c-d05c-4f0c-a17a-6c575fdd1ae0-0', usage_metadata={'input_tokens': 20, 'output_tokens': 7, 'total_tokens': 27})

print(ai_msg.content)

J'adore programmer.

Built-in tools

Gemini supports a range of tools that are executed server-side.

Google search

Info

Requires langchain-google-vertexai>=2.0.11

Gemini can execute a Google search and use the results to ground its responses:

from langchain_google_vertexai import ChatVertexAI

llm = ChatVertexAI(model="gemini-3.7-flash").bind_tools([{"google_search": {}}])

response = llm.invoke("What is today's news?")

Code execution

Info

Requires langchain-google-vertexai>=2.0.25

Gemini can generate and execute Python code:

from langchain_google_vertexai import ChatVertexAI

llm = ChatVertexAI(model="gemini-3.7-flash").bind_tools([{"code_execution": {}}])

response = llm.invoke("What is 3^3?")

API reference

For detailed documentation of all features and configuration options, head to the ChatVertexAI API reference.