Stoked seeing the OpenSearch Project featured by Jensen Huang on #NVIDIA #GTC keynote! ๐
One of the innovations in #OpenSearch V3 has been adding GPU acceleration based on NVIDIA's cuVS. Our #VectorSearch benchmarks, using CAGRA algorithm integrated through Facebook's Faiss library, showed:
โ
9.3x faster index builds
โ
3.75x lower cost
โ
2x higher throughput
โ
2.5x lower CPU usage
https://www.linkedin.com/feed/update/urn:li:activity:7439600547852189697/
#OpenSearchAmbassador #opensource #gtc2026 #gtc26 #cuvs #vectordb

#gtc #opensearch #nvidia #opensource #opensearchambassador #vectorsearch | Dotan Horovits
Stoked seeing the OpenSearch Project featured by Jensen Huang on NVIDIA #GTC keynote! ๐ One of the innovations in #OpenSearch V3 has been adding GPU acceleration based on #NVIDIA's cuVS. Our benchmarks, using CAGRA algorithm integrated through Facebook's Faiss library, showed: โ 9.3x faster index builds โ 3.75x lower cost โ 2x higher throughput โ 2.5x lower CPU usage That's the power of bringing the best of #opensource in vector search together. Check out the comments for the full benchmark setup and results, and more details on the architecture, as well as the RFC on GitHub. Well done to Navneet Verma Corey Nolet Kshitiz G. Dylan Tong Nathan Stephens Vamshi Vijay Nakkirtha and all involved! #OpenSearchAmbassador #VectorSearch