[Question]: is there a way to update rows in Qdrant vector store using llama index?

[mirallm]

Question Validation

• I have searched both the documentation and discord for an answer.

Question

https://docs.llamaindex.ai/en/stable/examples/vector_stores/QdrantIndexDemo/
Based on this doc, i cant find anything, but i would like to be able to update using llama index rows in qdrant especially node_content because it is actually created automaticall y with VectorStoreIndex @dosu

[dosubot]

I couldn't find specific methods in LlamaIndex for directly updating existing entries in a Qdrant vector store, particularly for updating the node_content. However, a common workaround is to delete the existing entries and then re-add them with the updated information. This approach is demonstrated in the LlamaIndex documentation for other vector stores, such as Redis, where entries are updated by deleting and re-adding them ^[1].

If you need to update the node_content, you can follow a similar process:

1. Delete the existing entry using its node ID.
2. Modify the document as needed.
3. Re-index the modified document using the VectorStoreIndex.

This method ensures that the updated content is stored in the vector store.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[logan-markewich]

delete and re-insert imo

[mirallm]

@logan-markewich but i want to avoid recreating the embeddings since only metadata are changed

[logan-markewich]

You can get the existing node and copy the embeddings i think? Get the node, updste the metadata, delete, add again

vector_store.get_nodes([node_id]) -> vector_store.delete_nodes([node_id]) -> vector_store.add([node])