A practical RAG case raises GraphRAG cost questions
A traditional RAG app built on docs ran into trouble during data preparation. Chunking failed, and manually setting overlapping chunks took about 49 minutes to ingest into a .
The app was completed, but the step returned answers that were not relevant enough. The likely problem was a mistake in how the source material was ingested and stored.
GraphDB and are being considered as alternatives, with the main question being when is useful, when it is unnecessary, and what it brings. There is also a cost concern if the graph keeps growing during each chat.
Key points
- The traditional RAG setup had problems during chunking and .
- Manual overlap handling took about 49 minutes before the data reached the .
- The finished app retrieved weak or unrelated answers.
- is being evaluated for its pros, cons, and right .
- A growing graph during chat could increase cost and processing time.