How to set up Google Vertex AI in BoltAI

Written By Daniel Nguyen

Last updated 21 days ago

Google Vertex AI is a unified, end-to-end machine learning (ML) platform provided by Google Cloud. It's designed to help data scientists and ML engineers build, deploy, and scale machine learning models more efficiently, from data ingestion to model monitoring, all within a single environment.

BoltAI supports Google Vertex AI as a native provider. Follow these quick steps to set up and chat with Vertex AI models from inside BoltAI.

Prerequisite

  • Make sure your Google Cloud project has billing enabled and the Vertex AI API turned on.

  • Create or choose a service account and grant it the Vertex AI User role (or Vertex AI Admin).

  • Create a JSON service account key for that service account in the Cloud Console (Service Accounts → Keys → Add key → Create new key → JSON). The key file is downloaded once and cannot be downloaded again, so store it securely.

  • Pick a Vertex AI location that supports the models/features you want to use; availability varies by region.

  • Optional: If you use the global endpoint, set location to global (note the global endpoint has limitations and no data residency guarantees).

Set up Google Vertex AI provider in BoltAI

  1. Open BoltAI → Settings → AI Services.

  2. Click Add → choose Google Vertex AI → Next.

  3. Under API Configuration, click Choose JSON File and select your service account key.

  4. Set Location (required). Choose a region from the list or pick Custom

  5. (Optional) Enter Project ID. If left blank, BoltAI can read it from your JSON key.

  6. Set a Default Model → Add Service.

Note that Gemini 3 Pro model is only available in global endpoint. Check out this guide for details.

Configure Google Vertex AI in BoltAI
Set default Vertex AI model

Start a new chat with Vertex AI models

  1. Click New Chat

  2. Choose the Google Vertex AI service/model in the model switcher.

  3. Optional) Enable native tools like Google Search, Code Execution or Enterprise Web Search…

  4. Send your message.

How to use Google Vertex AI in BoltAI

Available Tools

Code Execution

With Code Execution, certain Gemini models on Vertex AI can generate and execute Python code. This allows the model to perform calculations, data manipulation, and other programmatic tasks to enhance its responses.

URL Context

URL Context allows Gemini models to retrieve and analyze content from URLs. Supported models: Gemini 2.5 Flash-Lite, 2.5 Pro, 2.5 Flash, 2.0 Flash.

Google Search

Google Search enables Gemini models to access real-time web information. Supported models: Gemini 2.5 Flash-Lite, 2.5 Flash, 2.0 Flash, 2.5 Pro.

Enterprise Web Search

Enterprise Web Search provides grounding using a compliance-focused web index designed for highly-regulated industries such as finance, healthcare, and the public sector. Unlike standard Google Search grounding, Enterprise Web Search does not log customer data and supports VPC service controls. Supported models: Gemini 2.0 and newer.

Google Maps

Google Maps grounding enables Gemini models to access Google Maps data for location-aware responses. Supported models: Gemini 2.5 Flash-Lite, 2.5 Flash, 2.0 Flash, 2.5 Pro, 3.0 Pro.