NVIDIA CAGRA

NVIDIA CAGRA is a GPU-accelerated library designed for approximate nearest neighbor (ANN) searches, specifically optimized for high-performance vector retrieval using modern GPU parallelism.

Features

• GPU acceleration for graph-based approximate nearest neighbor search
• Optimized for high-performance vector search and large-scale data sets
• Leverages modern GPU parallelism for rapid vector retrieval
• Suitable for scenarios requiring fast and scalable vector search

Category

SDKs & Libraries

Tags

• gpu-acceleration
• ann
• high-performance
• vector-search

Pricing

No pricing information is provided in the available content.