Box Integrates Nvidia Nemotron Embeddings into Enterprise Content Retrieval Product
Box announced Box Content Retrieval, a new feature built on Nvidia's Nemotron Embed open-weight embedding models, aimed at improving AI agent search accuracy within enterprise document repositories. The feature leverages Nemotron's semantic embeddings alongside Box's existing metadata and permission controls to let agents retrieve relevant document chunks instead of parsing entire files. Nvidia claims the system delivers roughly four times faster queries, 20% fewer errors, and 90% fewer tokens consumed. The feature will roll out to Box Enterprise Advance customers starting early next year. This is a product/business announcement with no security vulnerability, incident, or threat activity involved.
Linha do tempo
-
Box Integrates Nvidia Nemotron Embeddings into Enterprise Content Retrieval Product
ler a análise desta fonte ocultar
Box announced Box Content Retrieval, a new feature built on Nvidia's Nemotron Embed open-weight embedding models, aimed at improving AI agent search accuracy within enterprise document repositories. The feature leverages Nemotron's semantic embeddings alongside Box's existing metadata and permission controls to let agents retrieve relevant document chunks instead of parsing entire files. Nvidia claims the system delivers roughly four times faster queries, 20% fewer errors, and 90% fewer tokens consumed. The feature will roll out to Box Enterprise Advance customers starting early next year. This is a product/business announcement with no security vulnerability, incident, or threat activity involved.