llama-index-embeddings-google-genai

What it is and what it does

This package wraps Google's embedding APIs (Gemini and Vertex AI) for use within the llama-index framework. It lets you generate text embeddings using Google's models without writing the integration layer yourself. The package handles both direct GenAI API calls and Vertex AI deployments, accepting configuration like project ID and location for the latter.

You instantiate an embedding model by specifying the model name (e.g., "gemini-embedding-2-preview") and optional Vertex AI credentials, then call get_text_embedding() to convert text into vector form. It's designed as a drop-in embedding provider for llama-index workflows that need Google's embedding models.

Use it for:

• Building RAG pipelines with llama-index using Google's embedding models
• Generating embeddings for semantic search over document collections via Gemini or Vertex AI
• Integrating Google's embeddings into multi-step llama-index workflows without custom API glue code
• Using Vertex AI embeddings in enterprise GCP environments with project-scoped authentication

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Integrates Google's Gemini and Vertex AI embedding models with llama-index for generating text embeddings in your applications.

Yes, if you are already using llama-index and need Google's embedding models. The low install friction, active maintenance, MIT license, and zero known vulnerabilities make it a straightforward choice. No if you don't use llama-index or prefer other embedding providers.

Install

llama-index-embeddings-google-genai on PyPI

pip install llama-index-embeddings-google-genai

uv add llama-index-embeddings-google-genai

poetry add llama-index-embeddings-google-genai

Installing llama-index-embeddings-google-genai

Before you install

Low install friction with only 2 runtime dependencies (google-genai and llama-index-core). Maintenance status is active, with the latest release dated 2026-05-19.

License in practice

MIT license permits free use, modification, and distribution with minimal restrictions, making it suitable for both open-source and commercial projects.

Quickstart

pip install llama-index-embeddings-google-genai

from llama_index.embeddings.google_genai import GoogleGenAIEmbedding

embed_model = GoogleGenAIEmbedding(model_name="gemini-embedding-2-preview")
embeddings = embed_model.get_text_embedding("Hello, world!")

Requires Python 3.10 or later (supports_current). Google GenAI API credentials must be configured.

Verify before relying

• Whether google-genai dependency requires API credentials or authentication setup beyond standard environment variables
• Performance characteristics and rate limits when used with Vertex AI vs. direct GenAI API
• Compatibility guarantees with specific versions of llama-index-core

Package facts

Evidence: llama_index_embeddings_google_genai-0.5.1-py3-none-any.whl

Tags

More Artificial Intelligence packages