AI Briefing#gb300
The latest AI and developer news about #gb300, with the original source and a short summary.
The latest AI and developer news about #gb300, with the original source and a short summary.
By optimizing the memory hierarchy across GPUs, CPUs, and storage, NVIDIA achieves a 30x increase in agentic throughput per dollar with only a 3% rise in HBM capacity.
The New Albany facility supports Muse model training with over 90% RL goodput and aims for a 100k GPU-hour mean time between interruptions.
SemiAnalysis data shows NVIDIA Vera Rubin NVL72 systems deliver over 30x higher throughput per megawatt than GB300 NVL72.
NVIDIA Vera Rubin NVL72 took the lead in MLPerf Inference v6.1, recording up to 3.7x the throughput of the GB300.
HP has started taking orders for the ZGX Fury, an enterprise AI inference workstation featuring the NVIDIA GB300 Grace Blackwell Ultra superchip and 748GB of unified memory.
Cursor released an open-source MoE megakernel for NVL72, boosting throughput by up to 2.37x.
Anthropic's Claude models are now generally available on Microsoft Azure powered by NVIDIA GB300 Blackwell Ultra GPUs.
SpaceX has nearly completed development of its own C-based AI training stack V1.0, targeting a performance improvement of more than 10x over JAX.
NVIDIA has automated its development and research workflows with Codex and GPT-5.5.