Box builds agent-oriented content retrieval system on NVIDIA Nemotron, claims 4x faster answers and 90% fewer tokens (report)

Verified (fact)

According to TechCrunch, Box launched a new content retrieval system built on NVIDIA Nemotron models and designed primarily for use by AI agents. The baseline for comparison, measurement conditions, and availability are not in the collected excerpt.

Needs fact-check (claim)

Query responses were described as 4x faster, with 20% fewer errors and 90% fewer tokens used (TechCrunch used the word 'reportedly').

Why it matters (interpretation)

An enterprise document storage company has designed retrieval specifically for agents rather than for people. In the agent era, the step of pulling internal documents is emerging as a key stage that determines cost and accuracy.

Sources

• Press Box taps Nvidia’s Nemotron for new agent-first content retrieval system TechCrunch