TokenPost topic
GB300
4 TokenPost stories are tagged GB300. The newest appear first.
Latest GB300 stories
- Nvidia Says Vera Rubin Reaches Up to 30 Times GB300’s Cost-Adjusted Inference Throughput
Vera Rubin NVL72’s per-GPU high-bandwidth memory (HBM) capacity is about 3% higher than GB300 NVL72’s. GB300 NVL72 reached up to 10 times H200’s cost-adjusted inference throughput on specific workloads.
· Technology
- Musk Targets 200,000 Monthly Chip Additions in SpaceX AI Buildout
He expects about 10 gigawatts of additional AI compute online in 2027, beyond the 2.5 to 3 gigawatts targeted for the end of 2026.
· Technology
- NVIDIA Tests Robots to Assemble Tester Trays for GB300 AI Systems
The experimental setup handles busbar fastening and four-connector insertion, but the systems have not yet reached factory deployment.
· Technology
- xAI’s Colossus 2 Could Reach 1.21 Million Chips by Year-End
The cluster currently has 550,000 GB200 and GB300 chips, with as many as 660,000 additional GB300 chips potentially coming online.
· Technology