Ai2 replaces priority scheduler with GPU time budgets

Ai2 replaced priority-based scheduling with GPU time budgets, hierarchical fair-share allocation, and time slicing, shifting allocation from case-by-case debates to a transparent budgeting process. Its clusters host thousands of NVIDIA H100, B200, and B300 GPUs for about 150 researchers, with demand often 2-3x supply.

• Economics of Open-Weight Inference: Older GPUs Retain Value2026-09-22
• Exposed Nvidia GPU monitors can reveal AI infrastructure secrets2026-10-09
• Nvidia-Backed Lumentum Sees Opto-Parts Sold Out to 20292026-10-09
• Trump unveils $2.4B in tech giant AI pledges2026-10-09
• Developers build simulations with frontier AI agents and Omniverse2026-10-08