← all datasets

cuVS Bench

Emerging
1papers using it
2026first seen

The 'cuVS Bench' is a benchmark dataset used to evaluate the performance of GPU-native approximate nearest neighbor search algorithms, specifically focusing on metrics such as index build time, search throughput, recall, and storage efficiency.

🔎 Find this dataset

Papers using cuVS Bench (1)

cuVS Bench — datasets — learning-to-hash