Mixed-precision numerics in scientific applications: survey and perspectives

The explosive demand for artificial intelligence (AI) workloads has led to a significant increase in silicon area dedicated to lower-precision computations on recent high-performance computing hard…

v (@vsoch.bsky.social)

We are excited to announce support for Flux for #Kubeflow v2.2 to enable AI/ML workloads paired with #HPC simulation. Flux adds a ZeroMQ bootstrap, support for #PMIx, more flavors of #MPI, and bypasses potential etcd and kube-sched bottlenecks. We are excited to bring this to the larger community! 🥳 [contains quote post or other embedded content]

Price of Nvidia's Vera Rubin NVL72 racks skyrockets to as much as $8.8 million apiece, but server makers' margins will be tight — Nvidia is moving closer to shipping entire full-scale systems

Nvidia and other chipmakers will still make plenty of cash.

Snakemake

Scalable Training of Mixture-of-Experts Models with Megatron Core

Scaling Mixture-of-Experts (MoE) training introduces systems challenges absent in dense models. Because each token activates only a subset of experts, this sparsity allows total parameters to grow much faster than per-token computation, creating coupled constraints across memory, communication, and computation. Optimizing one dimension often shifts pressure to another, demanding co-design across the full system stack. We address these challenges for MoE training through integrated optimizations spanning memory (fine-grained recomputation, offloading, etc.), communication (optimized dispatchers, overlapping, etc.), and computation (Grouped GEMM, fusions, CUDA Graphs, etc.). The framework also provides Parallel Folding for flexible multi-dimensional parallelism, low-precision training support for FP8 and NVFP4, and efficient long-context training. On NVIDIA GB300 and GB200, it achieves 1,233/1,048 TFLOPS/GPU for DeepSeek-V3-685B and 974/919 TFLOPS/GPU for Qwen3-235B. As a performant, scalable, and production-ready open-source solution, it has been used across academia and industry for training MoE models ranging from billions to trillions of parameters on clusters scaling up to thousands of GPUs. This report explains how these techniques work, their trade-offs, and their interactions at the systems level, providing practical guidance for scaling MoE models with Megatron Core.