Skip to content

Google Vertex

Overview

Google Vertex AI is a managed ML platform that provides access to Gemini, PaLM, and third-party models through Google Cloud. Floopy supports Vertex AI via custom base URL configuration, allowing you to route requests through your own Vertex AI endpoint.

For direct Gemini API access (without Vertex AI), see the Gemini provider guide.

Setup

Vertex AI requires a custom base URL because each GCP project has a unique endpoint based on project ID and region.

  1. Go to Settings > Providers in the dashboard.
  2. Click Add provider and select Google Vertex.
  3. Enter your Vertex AI endpoint URL (e.g., https://us-central1-aiplatform.googleapis.com/v1/projects/YOUR_PROJECT/locations/us-central1).
  4. Paste your GCP service account credentials and click Save.

Supported Models

Models available depend on your GCP project’s Vertex AI model access. Common models include:

ModelNotes
gemini-2.5-proGemini 2.5 Pro via Vertex
gemini-2.0-flashGemini 2.0 Flash via Vertex
gemini-1.5-proGemini 1.5 Pro via Vertex

Usage

import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.floopy.ai/v1",
apiKey: process.env.FLOOPY_API_KEY,
});
const response = await client.chat.completions.create({
model: "gemini-2.5-pro",
messages: [{ role: "user", content: "Explain quantum computing." }],
});

Provider-Specific Features

  • Custom base URL — Your Vertex AI endpoint is configured per-organization in provider settings.
  • GCP integration — Leverage existing GCP IAM, VPC Service Controls, and compliance configurations.
  • Region selection — Choose the GCP region closest to your infrastructure for lowest latency.
  • Vertex vs Gemini API — Use Vertex when you need GCP-native security controls and billing. Use the standard Gemini provider for direct API access.