run-llama/llama_index · error · NotImplementedError

Cannot delete from an empty index.

Error message

Cannot delete from an empty index.

What it means

EmptyIndex stores nothing, so node deletion is meaningless and _delete_node() is a hard NotImplementedError stub. It fires whenever the base BaseIndex delete machinery (delete_ref_doc / delete_nodes with ref_doc_id or node ids) dispatches to this method.

Source

Thrown at llama-index-core/llama_index/core/indices/empty/base.py:86

        Args:
            documents (List[BaseDocument]): A list of documents.

        Returns:
            IndexList: The created summary index.

        """
        del nodes  # Unused
        return EmptyIndexStruct()

    def _insert(self, nodes: Sequence[BaseNode], **insert_kwargs: Any) -> None:
        """Insert a document."""
        del nodes  # Unused
        raise NotImplementedError("Cannot insert into an empty index.")

    def _delete_node(self, node_id: str, **delete_kwargs: Any) -> None:
        """Delete a node."""
        raise NotImplementedError("Cannot delete from an empty index.")

    @property
    def ref_doc_info(self) -> Dict[str, RefDocInfo]:
        """Retrieve a dict mapping of ingested documents and their nodes+metadata."""
        raise NotImplementedError("ref_doc_info not supported for an empty index.")


# legacy
GPTEmptyIndex = EmptyIndex

View on GitHub (pinned to afd0fef371)

Solutions

1. Skip delete calls for EmptyIndex: check isinstance(index, EmptyIndex) before invoking delete_ref_doc/delete_nodes.
2. Branch on whether the index actually holds data (e.g. index.ref_doc_info is unsupported too, so track emptiness yourself) and route deletes only to data-bearing index types.
3. If deletions are required, use SummaryIndex or VectorStoreIndex, which implement _delete_node.

Example fix

// before
empty_index.delete_ref_doc(doc_id)  # NotImplementedError

// after
from llama_index.core.indices.empty import EmptyIndex
if not isinstance(index, EmptyIndex):
    index.delete_ref_doc(doc_id)

from llama_index.core.indices.empty import EmptyIndex

if not isinstance(index, EmptyIndex):
    index.delete_ref_doc(doc_id)

from llama_index.core.indices.empty import EmptyIndex

def supports_delete(index: object) -> bool:
    return not isinstance(index, EmptyIndex)

try:
    index.delete_nodes(node_ids)
except NotImplementedError as e:
    if "empty index" in str(e):
        pass  # nothing to delete in an empty index
    else:
        raise

When it happens

Trigger: Calling empty_index.delete_ref_doc(doc_id) or empty_index.delete_nodes(node_ids) on an EmptyIndex; generic document-lifecycle code (refresh/delete stale docs) run against an EmptyIndex selected by config.

Common situations: Document-sync jobs that delete removed docs from 'the current index'; mirroring an external data source where the index may legitimately be empty; multi-tenant setups where some tenants have no data and get an EmptyIndex.

Related errors

• Cannot insert into an empty index.
• ref_doc_info not supported for an empty index.
• Delete is not supported for KG index yet.
• _delete_node not implemented for BaseManagedIndex.
• Delete not implemented for tree index.

AI-assisted analysis of run-llama/llama_index@afd0fef371 (2026-08-15). Data as JSON: /api/errors/90eb13c392bf268a. Report an issue: GitHub.