Local Object Transport Accelerator throughput
Company: CoreWeave
The claim, verbatim
CoreWeave's Local Object Transport Accelerator delivers up to 7 GB/s of throughput per GPU
Source (primary)
CoreWeave Boosts Agentic AI With Multi-Rack NVIDIA Vera Rubin Clusters - Yahoo Finance (-, news_article)
View cached copy (2026-09-20)Live source ↗
Quote: “while delivering up to 7 GB/s of throughput per GPU”
How we checked this
Checked on September 25, 2026. The cited source supports every part of this claim.
The article reports up to 7 GB/s of throughput per GPU for the accelerator, as a figure stated by CoreWeave management.
Confirmed in the source:
- CoreWeave's Local Object Transport Accelerator delivers up to 7 GB/s of throughput per GPU, according to CoreWeave management
What we did: Read our cached copy of the publisher (https://finance.yahoo.com/technology/ai/articles/coreweave-boosts-agentic-ai-multi-120400139.html) in full (11,926 characters, retrieved September 20, 2026) and checked each assertion in the claim against it.
Additional evidence
confirms CoreWeave Boosts Agentic AI With Multi-Rack NVIDIA Vera Rubin Clusters - Yahoo Finance
Quote: “while delivering up to 7 GB/s of throughput per GPU”