#nvidia-h100

The latest AI and developer news about #nvidia-h100, with the original source and a short summary.

PyTorch adds compile-time heuristics to Helion for faster kernel defaults

PyTorch's Helion now uses compile-time heuristics to select fast kernel configurations without autotuning, achieving 1.18x speedup over Flash Linear Attention and 1.88x over torch.compile for matmuls on H100.

·2·2026.11.13 01:00

Dotwave Engine Serves 56 Concurrent Full-Duplex Voice Sessions on One H100, Cutting Costs by 98%

The new inference engine achieves 56 concurrent sessions on a single NVIDIA H100 SXM 80 GB by eliminating CPU-GPU overhead and batching continuous inference workloads.

·2·2026.10.02 22:00

What It Actually Takes to Run Code on a €200 Million Supercomputer

It explains the architecture of Spain's MareNostrum V supercomputer and how it actually operates using SLURM.

·2026.05.05 08:00

OpenAI Stargate: Status of US Sites

Stargate is scaling into a massive AI infrastructure network, deploying over 9GW across 7 sites in the US.

·1·2026.04.21 09:00

LLM Saves 3GB

Unweight reduces LLM size by 15~22% while maintaining accuracy.

·2026.04.18 16:00

HF-FriendliAI Model Deployment Integration

One-click model deployment via FriendliAI is now available on the Hugging Face Hub.

·2025.01.22 09:00