**ZenANN** is a high-performance approximate nearest neighbor (ANN) similarity search library designed to be user-friendly for Python developers. It provides multiple indexing methods, such as **IVF** (Inverted File Index), **HNSW** (Hierarchical Navigable Small World), and **hybrid-index structures** to balance between accuracy and speed. The computation kernel of ZenANN will be optimized for cache efficiency, SIMD acceleration, and algorithms enhancements beyond existing in-memory libraries.
0 commit comments