Models & Labs
Microsoft Unveils GridSFM for Power Grid Optimization
Microsoft Researchhigh confidence
Why it matters
- →GridSFM significantly reduces computation time for AC optimal power flow, enhancing grid efficiency.
- →It offers a versatile solution that generalizes across various grid topologies without retraining.
- →The model supports the integration of renewable energy, promoting sustainable grid operations.
Microsoft has introduced GridSFM, a new foundation model designed to optimize power grid operations by predicting AC optimal power flow in milliseconds. This model aims to address the computational challenges faced by traditional methods, which can take hours to solve. By providing rapid and accurate solutions, GridSFM could potentially save up to $20 billion annually in congestion costs. The model's ability to generalize across different grid topologies without retraining makes it a versatile tool for grid operators, enhancing both efficiency and the integration of renewable energy sources.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
What happened next
- 1Emerald AI Tests Flexible Data Center Power Management
MIT Technology Review AI · June 16, 2026 · Background
- 2
More in Models & Labs
Models & Labsmodels
llama.cpp b12029 adds CUDA 13 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally supporting NVIDIA's latest CUDA 13 stack alongside AMD's ROCm 10.0. The inclusion of native builds for Linux arm64 with Snapdragon chips, covering CPU, Adreno GPU, and Hexagon NPU, signals a serious push into mobile edge deployment. While Apple Silicon KleidiAI is disabled in this specific build, the broader hardware coverage means developers no longer need to compile from source for most modern accelerators. It’s less about new features and more about removing friction for anyone running local models on bleeding-edge or diverse hardware.
llama.cpp Releases
© The Verge AIModels & Labsother
