Find answers from the community

Updated 7 months ago

When I'm using `VectorStoreIndex(nodes,

At a glance

The community member is using VectorStoreIndex and has TextNodes with associated metadata. They are unsure whether just the text or both the text and metadata are being embedded. Additionally, they are encountering an error stating that the text is below the maximum token input count, but it is still exceeding the maximum count.

In the comments, another community member confirms that both the text and metadata are embedded, and suggests using node.get_content(metadata_mode="embed") to control what gets embedded. Another community member provides a link to the relevant documentation on advanced metadata customization.

There is no explicitly marked answer in the comments.

Useful resources
When I'm using VectorStoreIndex(nodes, embed_model=embed_model, storage_context=storage_context, show_progress=True) and the TextNodes have metadata associated with them, what gets embedded, just the text or text and metadata? I have some nodes where the text is below the maximum token input count but I'm getting an error stating that it exceeds the maximum count
Add a reply
Sign up and join the conversation on Discord