{"id":"tp-40939","type":"article-newspaper","title":"Vera Rubin Inference Results Leave Nvidia’s 75% Training Claim Untested","author":"Simon Yoon","publisher":"TOKENPOST Inc.","url":"https://www.tokenpost.com/news/technology/40939","datePublished":"2026-10-26T10:51:33.000Z","dateModified":"2026-10-26T10:51:33.000Z","language":"en","abstract":"At about 190 output tokens per second per user, Vera Rubin reached 5.7 times GB300’s total throughput per megawatt of GPU and memory power, using a four-GPU GB300 comparison.","section":"Technology","references":[{"title":"production rollout of the platform","url":"https://www.tokenpost.com/news/technology/25883"}]}