![[str_idx_hnsw.png|256]] - **Description**: A multi-layered graph structure for Approximate Nearest Neighbor (ANN) search. It provides the best query speed and recall (accuracy) trade-off for vector embeddings. - **Extension**: Requires `pgvector`. - **Supported Operators**: `<->` (L2 distance), `<#>` (inner product), `<=>` (cosine distance). - **Key Parameters**: - `m`: Max connections per node (default 16). - `ef_construction`: Size of the dynamic candidate list during graph building (default 64). - **Metaphor**: A multi-layered high-speed rail network. The top layers let you zoom across the "map" to the right region (Express), while the lower layers provide fine-grained navigation to the exact station (Local).