Parasail will deploy d-Matrix Corsair inference accelerators alongside NVIDIA Hopper and Blackwell GPUs in a heterogeneous inference system, targeting faster token generation for cloud customers. The …
NVIDIA released the Nemotron 3 Ultra NVFP4 checkpoint, a quantized model that achieves up to 5.9x higher inference throughput than GLM-5.1 754B FP4 on decode-heavy workloads while matching BF16 accura…
A developer detailed the architectural differences between NVIDIA's Ampere and Hopper GPU architectures, focusing on tensor core and memory bandwidth improvements. The Hopper architecture introduces T…
NVIDIA released Qwen3.6-35B-A3B-NVFP4, a post-training FP4-quantized variant of Alibaba's 35B MoE model that fits on a single H100 by reducing VRAM from ~71 GB to ~23 GB. The quantization targets weig…
Fortinet announced an expanded integration of its FortiAIGate solution with Nvidia's AI platforms on May 12, embedding zero-trust security directly into Nvidia's Blackwell and Hopper GPUs to protect a…
NVIDIA released a guide showing how to optimize transformer-based models for low-precision training using Hopper and Blackwell GPUs, focusing on FP8 and NVFP4 formats. The method translates model conf…
INT21 launched as the first company to achieve self-improving AI agent swarms for AI infrastructure, introducing PTX Kernel Factory that generates and optimizes low-level GPU software for NVIDIA hardw…
Nvidia's Blackwell architecture achieves 20 times more AI agents per megawatt than its Hopper generation, according to the new AgentPerf benchmark. The efficiency leap, driven by FP4 precision and adv…
NVIDIA's Blackwell Ultra NVL72 platform achieved top performance on the first agentic AI benchmark, AgentPerf from Artificial Analysis, running up to 20 times more agents per megawatt than the previou…
NVIDIA's Blackwell consumer GPU (GB203/SM_120) features a unified TensorCore pipeline where all 12 non-FP64 precision formats share identical 29-cycle latency and 23-cycle throughput, reducing precisi…
FlashLib, a new GPU library for classical machine learning operators, achieves speedups of up to 208× over cuML on Hopper GPUs for algorithms including KMeans, KNN, and PCA. The library is designed to…
NVIDIA released CUDA 13.3, introducing tile programming in C++ that automates low-level GPU management for optimized kernel development across all supported architectures. The update also includes CUD…
Nvidia's Hopper and Blackwell GPU architectures introduced spatial scheduling through warp specialization, requiring developers to divide pipeline work between different warp groups for data movement …
Nvidia's Blackwell B200 GPU is the company's first chiplet design, using two reticle-sized dies that appear as a single GPU to software, moving away from the monolithic approach used in prior generati…