Box 基于 Nvidia Nemotron 嵌入模型推出面向智能体的内容检索系统
Box taps Nvidia’s Nemotron for new agent-first content retrieval system
- 1Box 在开发者大会发布 Box Content Retrieval，将 Nvidia Nemotron 嵌入模型接入企业内容检索
- 2Nvidia 副总裁称查询预计提速 4 倍、错误减少 20%、token 用量减少 90%（官方口径，暂无独立验证）
- 3功能面向所有 Box Enterprise Advance 客户，明年初开始推出
- 4系统依赖 Box 既有元数据与权限控制，CEO 称把五种技术打包进一个 API
嵌入向量语义检索正成为企业智能体的基础设施，Box 用自家元数据与权限控制叠加开源嵌入模型，形成难以从零复制的组合。
- 影响：对要在内部文档上部署智能体的企业，检索错误率和 token 开销是直接成本项，嵌入检索正好对应这两处；内容平台选型时检索与权限绑定的深度影响更换成本，迁移难度会随使用加深。
- 参考：可考虑先算出当前文档智能体的查询错误率与 token 开销作基线，再对照嵌入检索的收益估算；已在 Box 上的团队值得核对 Enterprise Advance 档位范围与明年初的上线节奏。
- 依据：提速 4 倍、错误少 20%、token 少 90% 为 Nvidia 副总裁向 TechCrunch 给出的官方口径，暂无独立验证；7 月 Nemotron 登顶 RTEB 基准为报道转述。
At its developer conference on Thursday, Box showed off a new feature that will build Nvidia’s Nemotron embedding system into the core of its product.
Called Box Content Retrieval, the new feature allows agents to answer queries far more effectively by relying on the embedded metadata created by Nemotron.
Box says the upshot for users is faster, more accurate and less expensive queries. Nvidia’s VP of generative AI software, Kari Briski, told TechCrunch that she expects queries to be answered four times faster, with 20% fewer errors and 90% fewer tokens used.
The system will be available to all of Box’s Enterprise Advance customers, rolling out early next year.
“This dramatically boosts both the relevance of our search technology for agents,” Box CEO Aaron Levie said. “Instead of an agent having to kind of read through each document, we can send back chunks from the documents that are most relevant to the user’s query or whatever the task is that the agent is doing.”
The feature is an implementation of Nvidia’s Nemotron Embed project, which offers open-weight embedding models to improve automated retrieval systems. Major labs have often overlooked such models, which use LLMs to pre-process data so it can be better used by agents. The sector has been a focus for Nvidia’s open-source model project, with its recent Nemotron release in July topping the RTEB benchmark, which measures retrieval quality.
The system will be most useful when paired with active agents, Box says, giving the example of lawyers, who will ostensibly be able to find relevant clauses and precedents more easily under the new system. Because Nemotron’s embeddings go beyond simple keywords to map underlying concepts, a search tool can surface related legal themes even when different wording or language is used.
While the Nemotron models are freely available, the system also relies heavily on Box’s pre-existing metadata and permission controls, which will make it challenging to recreate from scratch. “We’re probably packing in five different technologies into one simple API,” Levie told TechCrunch.
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.
Russell Brandom has been covering the tech industry since 2012, with a focus on platform policy and emerging technologies. He previously worked at The Verge and Rest of World, and has written for Wired, The Awl and MIT’s Technology Review. He can be reached at [email protected] or on Signal at 412-401-5489.
来源：TechCrunch · AI · techcrunch.com