ai-agents-metadata-filtering-multiple-metadatafilters-with-condition-filtercond-efcbcae8

THEN Instead of relying on the built-in OR filter, perform separate queries for each metadata filter and manually merge the resulting nodes. This ensures all nodes matching any filter are retrieved.

from llama_index.vector_stores.types import MetadataFilters, MetadataFilter, FilterCondition

filters = [MetadataFilter(key='color', value='red'), MetadataFilter(key='alpha', value='1')]
nodes = []
for f in filters:
    mf = MetadataFilters(filters=[f], condition=None)
    query_engine = index.as_query_engine(filters=mf)
    nodes.extend(query_engine.retrieve('your query'))
# deduplicate nodes if needed