Overview

This node integrates with Databricks to generate embeddings for documents and queries using a specified embeddings serving endpoint. It is useful in scenarios where you need to convert text data into vector representations for tasks like semantic search, similarity comparison, or machine learning workflows. For example, it can be used to embed a set of documents to store in a vector database or to embed a query for similarity search against stored vectors.

Use Case Examples

1. Embedding a batch of documents to store in a vector store for semantic search.
2. Embedding a user query to find the most relevant documents based on vector similarity.

Properties

Output

JSON

• response - The embeddings response object containing the generated embeddings for the input documents or queries.

Dependencies

• Databricks API key credential for authentication
• Databricks host URL

Troubleshooting

• Ensure the Databricks API key and host credentials are correctly configured and valid.
• Verify that the serving endpoint name is correct and exists in the Databricks environment.
• Check that the vector store and embedding model have the same dimensionality to avoid mismatches.
• Common error: 'Databricks API error' indicates issues with the API request or credentials; verify the token and endpoint URL.
• Unexpected response format errors suggest the API response structure has changed or the request payload is incorrect.

Links

• Databricks Generative AI Embeddings Documentation - Official Databricks documentation on creating query vector search using embeddings.