The GB300 follows the B200 in NVIDIA’s Blackwell data-center line (product page); NVIDIA’s page presents it as the GB300 Grace Blackwell Ultra Superchip and, in the GB300 NVL72 rack, as 72 Blackwell Ultra GPUs paired with 36 Grace CPUs. In this corpus it appears as the training substrate of Grok 4.5, trained on tens of thousands of them alongside Cursor, the run whose gains Aditya Gupta credited to a loop where better behavior unlocks more intelligence.
Hardware
NVIDIA GB300
NVIDIA's GB300 data-center accelerator, the chip SpaceXAI trained Grok 4.5 on by the tens of thousands.
Sources
Attribution and reporting linked by this record; links do not imply independent verification.
- Welcome to July 9, 2026 (theinnermostloop.substack.com)
Development timeline
Newest reporting dates first. Reporting dates may differ from event dates.
Knowledge graphopen in browser →
Developed by
Cite or correct this record
Suggest a correction — opens a GitHub draft for you to review and submit.