#nvidia-b200

The latest AI and developer news about #nvidia-b200, with the original source and a short summary.

New Kernel Fusion Techniques Hide Up to 90% of Normalization Latency in GEMM and Attention Kernels

New kernel fusion techniques for LayerNorm and RMSNorm hide up to 90% of normalization latency in GEMM kernels and achieve up to 35% speedup in attention kernels on NVIDIA B200 GPUs.

·2026.11.14 21:00

OpenBMB AI Agent Beats NVIDIA FlashMLA Baseline by 9.8% on B200 via Black-Box Optimization

The agent optimized kernels against a black-box baseline without source code access, releasing the harness and 27 operators.

·2026.11.02 23:00

JSSA Achieves 6.43x Decoding Throughput on B200 GPUs for 128K Context

The method maintains near-full-KV accuracy while supporting 700B scale models at 96-99% of DSA throughput.

·2026.10.29 17:00

Naver Cloud Achieves 1.33x Speedup on Single Layer via Mamba-2 Kernel Optimization for B200 GPUs

Naver Cloud optimized the Mamba-2 kernel for B200 GPUs using Helion DSL, increasing single-layer computation speed by 1.33x.

·1·2026.09.04 09:00

Naver Cloud Selected for National Security AI Project

The Ministry of Science and ICT (MSIT) has selected the Naver Cloud consortium as the final executing agency for the project to develop a cybersecurity-specialized AI foundation model.

·2026.09.03 10:00

How to Build an Ultra-Fast API for GLM-5.2

Baseten has launched an ultra-fast API dedicated to GLM-5.2 that maximizes performance by minimizing latency.

·1·2026.07.27 09:00

The winner flips depending on the model

In an apples-to-apples FP8 comparison, the advantage between Ironwood and B200 reversed depending on the model.

·1·2026.04.18 06:00