Vectors vs Embeddings - are embeddings now obsolete?

API
1

Give the recent updates to the Assistants API, which now include the automatic use of Vector Stores when uploading files through file search:
“Adding a file to a vector_store automatically parses, chunks, embeds and stores the file in a vector database that’s capable of both keyword and semantic search.”

Other than exceeding the 10,000 file limit, what situations would I need to create my own “embeddings”? Are embeddings now obsolete?

https://platform.openai.com/docs/guides/embeddings/what-are-embeddings

2

Couple of points here.

This functionality is specific to Assistants and not the API in general.

Leaving that aside, I would think about it as follows. The current design of the vector store and the approach for retrieval still comes with several limitations as are outlined in the documentation:

Known Limitations
We have a few known limitations we’re working on adding support for in the coming months:

  • Support for modifying chunking, embedding, and other retrieval configurations.
  • Support for deterministic pre-search filtering using custom metadata.
  • Support for parsing images within documents (including images of charts, graphs, tables etc.)
  • Support for retrievals over structured file formats (like csv or jsonl).
  • Better support for summarization — the tool today is optimized for search queries.

Whether these limitations present an issue to you, really depends on your use case and whether it requires a certain degree of customization in your embedding and retrieval process. If the answer is yes, then you may want to consider alternative solutions.

1 Like