An engineering team has already pre-computed text embeddings using a proprietary internal model and stored the embedding vectors in a Delta table. They want to create a Mosaic AI Vector Search index from these pre-computed embeddings. Which index option should they use?

Question

Accepted Answer

B. Delta Sync Index with self-managed embeddings, pointing to the Delta table that contains the pre-computed embedding vectors.. The Delta Sync Index with self-managed embeddings is designed for teams that pre-compute embeddings using their own (possibly proprietary) model and store those vectors in the source Delta table. Databricks indexes the pre-computed vectors and keeps the index in sync with the Delta table as it is updated.

An engineering team has already pre-computed text embeddings using a proprietary internal model and stored the embedding vectors in a Delta table. They want to create a Mosaic AI Vector Search index from these pre-computed embeddings. Which index option should they use?

Related Questions

Discussion