Box Integrates Nvidia’s Nemotron for Enhanced Content Retrieval
Box has announced a new feature called Box Content Retrieval, which integrates Nvidia’s Nemotron embedding system to improve query responses. This system is expected to enhance the speed and accuracy of content retrieval for users, particularly benefiting enterprise clients. The rollout is planned for early next year, promising significant improvements in how agents access and utilize information.
1 min read1 source
Key takeaways
30-second read- 1Developers can leverage the new Box Content Retrieval feature to create more efficient applications that rely on accurate and fast data retrieval.
- 2Product teams can enhance user experience by integrating this technology, which reduces errors and speeds up query responses.
- 3Operators in legal and other professional fields can expect improved tools for accessing relevant information, potentially increasing productivity.
Confirmed
No official source linked yet
Reported
Secondary reporting: TechCrunch AI
Verify yourself
Eligibility, local prices and dates for your car, state or utility

Box has introduced a new feature, Box Content Retrieval, which incorporates Nvidia’s Nemotron embedding system to enhance the way agents respond to queries. This integration aims to provide users with faster and more accurate search results by utilizing embedded metadata generated by Nemotron. According to Nvidia’s VP of generative AI software, Kari Briski, the new system is expected to answer queries four times faster, with a reduction of 20% in errors and a significant decrease in token usage by 90%.
The feature will be available to all Box Enterprise Advance customers starting early next year. Box CEO Aaron Levie emphasized that this advancement allows agents to retrieve relevant document segments without needing to sift through entire documents. Instead, the system can deliver specific chunks that directly address user queries, improving the relevance of search results.
The Nemotron embedding models, which are part of Nvidia’s open-source project, enhance automated retrieval systems by pre-processing data for better agent utilization. This is particularly beneficial in fields like law, where agents can quickly find pertinent clauses and precedents, even when different terminology is used. However, the effectiveness of this system also relies on Box’s existing metadata and permission controls, making it complex to replicate independently. Overall, this integration represents a significant step forward in content retrieval technology.
Sources
1 media- [1]Box taps Nvidia’s Nemotron for new agent-first content retrieval systemMedia
TechCrunch AI · techcrunch.com · published Nov 5
Newsletter
Weekly AI signal. No spam.




Comments
Sign in to join the discussion
Loading comments…