Google Gemini Embeddings

Generate embeddings using Google Gemini API

Overview

Google Gemini provides embedding models through the Gemini API with simple API key authentication.

Quick Start

from praisonaiagents import embedding

result = embedding(
    input="Hello world",
    model="gemini/text-embedding-004"
)
print(f"Dimensions: {len(result.embeddings[0])}")

CLI Usage

praisonai embed "Hello world" --model gemini/text-embedding-004

Setup

export GEMINI_API_KEY="your-gemini-api-key"
# or
export GOOGLE_API_KEY="your-google-api-key"

Available Models

Model	Dimensions	Use Case
gemini/text-embedding-004	768	Latest, recommended
gemini/embedding-001	768	Legacy model

Batch Embeddings

from praisonaiagents import embedding

texts = ["Document 1", "Document 2", "Document 3"]
result = embedding(
    input=texts,
    model="gemini/text-embedding-004"
)
print(f"Generated {len(result.embeddings)} embeddings")

Task Types

from praisonaiagents import embedding

# For retrieval queries
result = embedding(
    input="What is machine learning?",
    model="gemini/text-embedding-004",
    task_type="RETRIEVAL_QUERY"
)

# For documents
result = embedding(
    input="Machine learning is...",
    model="gemini/text-embedding-004",
    task_type="RETRIEVAL_DOCUMENT"
)

Related

• Embedding Providers Overview
• Vertex AI Embeddings