Where and how to store the summary_index?

Short Course Q&A Building Agentic RAG with LlamaIndex
1

If I have hundreds of documents, I’ll want to store my indexes and not re-create them each time I run my app. For a VectorStoreIndex, I know to store my embeddings in a vector database like Pinecone. I’m not familiar with SummaryIndex. Where and how would I store those?

2

Hello Sharonm,

Did you get any solution for it?

3

Hi @sharonm and @Muhammad_Haseeb1

Probably the below link will help you understand on how make a summary index

https://docs.llamaindex.ai/en/stable/examples/index_structs/doc_summary/DocSummary/

Regards
DP

4

Thanks, but the issue isn’t how to make a summary index. That was clear from the code. It’s how to STORE it and re-load / re-use it after it’s created.

5

I figured out the summary index is different in terms of structure, it means it’s not good to be stored in a vector database, but you can persist them directly to the disk and load from there.