How to beginner · 3 min read

How to use service account with Vertex AI

Q: How to use service account with Vertex AI

To use a service account with Vertex AI, set the GOOGLE_APPLICATION_CREDENTIALS environment variable to the path of your service account JSON key file. Then initialize the vertexai SDK in Python, which will automatically use these credentials for authentication.

Quick answer

To use a service account with Vertex AI, set the GOOGLE_APPLICATION_CREDENTIALS environment variable to the path of your service account JSON key file. Then initialize the vertexai SDK in Python, which will automatically use these credentials for authentication.

PREREQUISITES

Python 3.8+
Google Cloud project with Vertex AI enabled
Service account JSON key file with Vertex AI permissions
pip install vertexai google-cloud-aiplatform

Setup

Install the required Python packages and set the environment variable for your service account key.

Install the vertexai and google-cloud-aiplatform packages.
Download your service account JSON key from Google Cloud Console with Vertex AI permissions.
Set the GOOGLE_APPLICATION_CREDENTIALS environment variable to the key file path.

bash

pip install vertexai google-cloud-aiplatform

# On Linux/macOS
export GOOGLE_APPLICATION_CREDENTIALS="/path/to/your/service-account.json"

# On Windows (PowerShell)
$env:GOOGLE_APPLICATION_CREDENTIALS="C:\path\to\your\service-account.json"

Step by step

This example shows how to initialize the vertexai SDK using the service account credentials and generate a text completion with the gemini-2.0-flash model.

python

import os
import vertexai
from vertexai.generative_models import GenerativeModel

# Initialize Vertex AI SDK (auto-uses service account credentials)
vertexai.init(
    project=os.environ.get("GOOGLE_CLOUD_PROJECT"),
    location="us-central1"
)

# Load the Gemini model
model = GenerativeModel("gemini-2.0-flash")

# Generate text
response = model.generate_content("Explain how to use a service account with Vertex AI.")
print(response.text)

output

Explain how to use a service account with Vertex AI.

A service account allows secure authentication to Google Cloud services including Vertex AI. Set the environment variable GOOGLE_APPLICATION_CREDENTIALS to your service account JSON key file path. Then initialize the Vertex AI SDK in Python, which will use these credentials automatically.

Common variations

You can also authenticate using google.auth libraries or use asynchronous calls with the vertexai SDK. For streaming or chat models, use the respective classes like GenerativeModel with streaming or chat methods.

python

import asyncio
import os
import vertexai
from vertexai.generative_models import GenerativeModel

vertexai.init(
    project=os.environ.get("GOOGLE_CLOUD_PROJECT"),
    location="us-central1"
)

async def main():
    model = GenerativeModel("gemini-2.0-flash")
    response = model.generate_content("Hello, how do I use a service account with Vertex AI?")
    print(response.text)

asyncio.run(main())

output

Hello! To use a service account with Vertex AI, set the GOOGLE_APPLICATION_CREDENTIALS environment variable to your service account key file path. The Vertex AI SDK will then authenticate automatically using those credentials.

Troubleshooting

If you get authentication errors: Verify the GOOGLE_APPLICATION_CREDENTIALS path is correct and the service account has the necessary Vertex AI roles.
If the project or location is not set: Ensure you pass project and location to vertexai.init() or set them in your environment variables.
Permission denied errors: Check IAM permissions for the service account in Google Cloud Console.

✅

Key Takeaways

Set the GOOGLE_APPLICATION_CREDENTIALS env var to your service account JSON key file path.
Initialize the vertexai SDK with your Google Cloud project and location to authenticate automatically.
Use vertexai.generative_models classes like GenerativeModel for inference.
Verify IAM permissions and environment variables if authentication fails.

Verified 2026-04 · gemini-2.0-flash

Verify ↗