#nvl72

The latest AI and developer news about #nvl72, with the original source and a short summary.

SkyPilot Topology-Aware Scheduler Boosts NVL72 Efficiency by 27% via Rack-Aware Placement

A new scheduling strategy keeps workloads within NVLink domains to avoid 43% performance loss, yielding 27% more usable GPUs and 14.4% higher throughput in pre-training.

·1·2026.10.29 09:00

Cursor Releases MoE Kernel for NVL72

Cursor released a fused MoE communication and computation kernel for NVL72, reporting a 1.41x throughput improvement.

·2·2026.08.07 12:00

Mixture-of-Kittens: Open-Source MoE Megakernel for NVL72

Cursor released an open-source MoE megakernel for NVL72, boosting throughput by up to 2.37x.

·2026.08.05 09:00

NVIDIA Vera Rubin, Delivering Improved Performance per Watt and the Lowest Token Cost for Partners

NVIDIA begins mass production of Vera Rubin NVL72 and expands supply to global cloud partners.

·1·2026.07.22 00:00

PickWhy 'Performance per Watt' Is the Ultimate Metric for AI Infrastructure Efficiency

NVIDIA highlights performance per watt as the key metric determining the profitability of AI infrastructure in power-constrained environments, showcasing the overwhelming efficiency of the Blackwell platform.

·2·2026.07.15 00:00