Google Vertex AI Embeddings

Generate embeddings using Google Cloud Vertex AI

Overview

Google Vertex AI provides enterprise-grade embedding models including textembedding-gecko and multimodal embeddings.

Quick Start

from praisonaiagents import embedding

result = embedding(
    input="Hello world",
    model="vertex_ai/textembedding-gecko"
)
print(f"Dimensions: {len(result.embeddings[0])}")

CLI Usage

praisonai embed "Hello world" --model vertex_ai/textembedding-gecko

Setup

export GOOGLE_APPLICATION_CREDENTIALS="/path/to/service-account.json"
export VERTEXAI_PROJECT="your-project-id"
export VERTEXAI_LOCATION="us-central1"

Available Models

Model	Dimensions	Use Case
vertex_ai/textembedding-gecko	768	General purpose
vertex_ai/textembedding-gecko@003	768	Latest version
vertex_ai/textembedding-gecko-multilingual	768	Multilingual
vertex_ai/text-embedding-004	768	Newest model
vertex_ai/text-multilingual-embedding-002	768	Multilingual v2

With Project Configuration

from praisonaiagents import embedding

result = embedding(
    input="Hello world",
    model="vertex_ai/textembedding-gecko",
    vertex_project="my-project",
    vertex_location="us-central1"
)

Batch Embeddings

from praisonaiagents import embedding

texts = ["Document 1", "Document 2", "Document 3"]
result = embedding(
    input=texts,
    model="vertex_ai/textembedding-gecko"
)
print(f"Generated {len(result.embeddings)} embeddings")

Multimodal Embeddings

from praisonaiagents import embedding

# Image + text embeddings
result = embedding(
    input="A beautiful sunset",
    model="vertex_ai/multimodalembedding",
    image="base64_encoded_image_data"
)

Related

• Embedding Providers Overview
• Gemini Embeddings