Google Embeddings
Google Embeddings
Section titled “Google Embeddings”Leverage Google (Gemini) to create embeddings.
What it does
Section titled “What it does”Converts text into vector embeddings using a Google (Gemini) embeddings model. Google exposes an OpenAI-compatible Gemini API, so this node runs the OpenAI embeddings client against Google’s base URL.
When to use it
Section titled “When to use it”Connect this to indexer or RAG nodes when you need to create embeddings for semantic search and retrieval using Google (Gemini).
Inputs and settings
Section titled “Inputs and settings”| Setting | Notes |
|---|---|
| Model | The Google embeddings model to use, e.g. text-embedding-004. Placeholder: text-embedding-004. |
| Authentication | Credential field. Use your Google AI (Gemini) API key as a Bearer token. |
Outputs
Section titled “Outputs”Returns an embeddings dependency for vector stores and retrieval workflows.
Dependencies and credentials
Section titled “Dependencies and credentials”- Requires an Authentication credential (Google AI / Gemini API key), sent as a Bearer token.
- The node throws an error if Model is empty.
Example workflow
Section titled “Example workflow”Connect Google Embeddings to an AI agent or dependency input that accepts this dependency type, set the Model to a Google embeddings model such as text-embedding-004, attach your Google AI (Gemini) API key credential, then run the agent with data from previous nodes.
Troubleshooting
Section titled “Troubleshooting”- Check that the Model field is set — the node throws an error if it is empty.
- Verify the Google AI (Gemini) API key credential is valid.
- If it calls an external service, verify credentials, permissions, and rate limits.