Vector search looks for similar vectors or data points in a dataset based on their vector representations. However, pure vector search is rarely sufficient in real-world scenarios. Vectors usually come with metadata, and users often need to apply one or more filters to this metadata. That makes filtered vector search come into play.
Filtered vector search is becoming increasingly vital for intricate retrieval scenarios. You can apply a filtering mechanism to filter out the undesired vectors beyond the top-k/range of multi-dimensional embeddings.