@fixtse/n8n-nodes-google-gemini-embeddings-extended

This is an n8n community sub-node that provides Google Gemini Embeddings with extended features, including support for output dimensions, task types, and special handling for models like gemini-embedding-001.

Features

• Support for any Google Gemini embedding model (specify by name)
• Output dimensions configuration (for supported models)
• Task type specification for optimized embeddings
• Title support for retrieval documents
• Batch size control for rate limit management
• Special handling for gemini-embedding-001 (single input per request)
• Uses standard Google API credentials (same as other Google AI nodes)
• Works as a sub-node with vector stores and other AI nodes

Installation

Community Node (Recommended)

1. In n8n, go to Settings > Community Nodes
2. Search for @fixtse/n8n-nodes-google-gemini-embeddings-extended
3. Click Install

Manual Installation

npm install @fixtse/n8n-nodes-google-gemini-embeddings-extended

Setup

Prerequisites

1. A Google AI Studio account
2. A Gemini API key

Authentication

This node uses the standard Google PaLM/Gemini API credentials:

1. Get your API key from Google AI Studio
2. In n8n, create Google PaLM API credentials
3. Enter your API key

Usage

This is a sub-node that provides embeddings functionality to other n8n AI nodes.

Using with Vector Stores

1. Add a vector store node to your workflow (e.g., Pinecone, Qdrant, Supabase Vector Store)
2. Connect the Embeddings Google Gemini Extended node to the embeddings input of the vector store
3. Configure your Google PaLM API credentials
4. Enter your model name (e.g., text-embedding-004, gemini-embedding-001)
5. Configure additional options as needed
6. The vector store will use these embeddings to process your documents

Example Workflow

[Document Loader] → [Vector Store] ← [Embeddings Google Gemini Extended]
                          ↓
                    [AI Agent/Chain]

Configuration Options

Model Name

Enter any valid Google Gemini embedding model name. Examples:

• text-embedding-004 (Latest, supports output dimensions)
• gemini-embedding-001 (Supports output dimensions, processes one input at a time)
• embedding-001 (Legacy model)

Output Dimensions

For models that support it, you can specify the number of output dimensions:

• Set to 0 to use the model's default dimensions
• Set to a specific number (e.g., 256, 768, 3072) to get embeddings of that size

Task Types

Optimize your embeddings by specifying the task type:

• Retrieval Document: For document storage in retrieval systems
• Retrieval Query: For search queries
• Semantic Similarity: For comparing text similarity
• Classification: For text classification tasks
• Clustering: For grouping similar texts
• Question Answering: For Q&A systems
• Fact Verification: For fact-checking applications
• Code Retrieval Query: For code search

Additional Options

• Title: Add a title to documents (only for RETRIEVAL_DOCUMENT task type)
• Strip New Lines: Remove line breaks from input text (enabled by default)
• Batch Size: Control how many texts are processed at once (default: 100)

Use Cases

• Semantic Search: Generate embeddings for documents and queries in vector stores
• RAG Applications: Build retrieval-augmented generation systems
• Document Similarity: Find similar documents in your vector database
• Multi-language Support: Use models that support multiple languages
• Code Search: Use CODE_RETRIEVAL_QUERY for searching code repositories

Model-Specific Notes

gemini-embedding-001

This model has special requirements:

• Only accepts one text input per request
• The node automatically handles this limitation
• Processing may be slower for large datasets
• Supports output dimensions up to 3072

text-embedding-004

• Supports batch processing
• Default dimensions: 768
• Good balance of performance and quality

Differences from Official n8n Node

This community node extends the official Google Gemini Embeddings node with:

1. Output Dimensions Support: Configure the size of embedding vectors
2. Extended Task Types: More task type options for optimization
3. Title Support: Add titles to documents for better retrieval
4. Batch Size Control: Manage rate limits effectively
5. Better Error Messages: More detailed error information

Compatible Nodes

This embeddings node can be used with:

• Simple Vector Store
• Pinecone Vector Store
• Qdrant Vector Store
• Supabase Vector Store
• PGVector Vector Store
• Milvus Vector Store
• MongoDB Atlas Vector Store
• Zep Vector Store
• Question and Answer Chain
• AI Agent nodes

Troubleshooting

Common Issues

1. Authentication Errors

   • Ensure your Google PaLM API key is valid
   • Check that the API is enabled in your Google Cloud project
   • Verify you have sufficient quota

2. Model Errors

   • Verify the model name is spelled correctly
   • Check Google's documentation for valid model names

3. Rate Limit Errors

   • Reduce the batch size in options
   • Add delays between requests if processing large datasets

4. Dimension Errors

   • Not all models support custom dimensions
   • Check model documentation for supported dimension values

5. Bad Request Errors

   • gemini-embedding-001 only accepts one input at a time (handled automatically)
   • Ensure text inputs are within token limits

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT

Support

For issues and feature requests, please use the GitHub issue tracker.

Changelog

0.1.0

• Initial release
• Support for Google Gemini embeddings via API
• Output dimensions configuration
• Task type selection with extended options
• Title support for documents
• Batch size control
• Special handling for gemini-embedding-001