hardware

Scaling AI Inference Across Multiple GPUs Using NVIDIA TensorRT with Multi-Device Inference Support

NVIDIA TensorRT's multi-device inference feature scales inference across multiple GPUs, enabling deployment of models that exceed single-GPU memory or reducing latency for memory-bound workloads. Each

Hardware

The Ultimate Summer Sale Pairing: Steam Sale Meets GeForce NOW Discounts

DGX agent

Summer savings are heating up. From the Steam Summer Sale to GeForce NOW membership discounts, this week’s GFN Thursday delivers double the deals and more ways to get the most value from cloud gaming.

hardwarenvidia-blog

25 Jun 2026

The US awards $250M in CHIPS funds to a startup co-founded by billionaire mining magnate Robert Friedland to develop silicon-carbide chips and pulsed-power tech (James Attwood/Bloomberg)

James Attwood / Bloomberg: The US awards 250M in CHIPS funds to a startup co-founded by billionaire mining magnate Robert Friedland to develop silicon-carbide chips and pulsed-power tech — I-Pulse Inc

US Grid Constraints: Towards 40GW+ of Behind-The-Meter Datacenter by 2028?

This analysis examines the growing power demands of behind-the-meter datacenters in the US, projecting capacity could exceed 40GW by 2028, and discusses how grid constraints and infrastructure limitat

hardwaresemianalysis

Without a doubt: A world running foundation models on a billion robots won't be doing it locally. None of the embedded compute will look lik…

Without a doubt: A world running foundation models on a billion robots won't be doing it locally. None of the embedded compute will look like it does today. All of the embedded compute will be wireles

hardwaredylan-patel--x

Accelerating Disaggregated RL for Visual Generative LLMs with Diffusion-Based Parallelism and Trainer-Assisted Generation

arXiv:2606.24369v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregr

hardwarearxiv-cs-ai

Accelerating Transformers Fine-Tuning with NVIDIA NeMo AutoModel

NVIDIA NeMo AutoModel is a tool designed to accelerate the fine-tuning process of transformer models by automating model selection and configuration. The solution leverages NVIDIA's NeMo framework to

hardwarehugging-face

Advancing AI in Property Insurance with AMD and Vultr

This article discusses a collaboration between AMD and Vultr to implement AI solutions in the property insurance sector, leveraging AMD's processors and Vultr's cloud infrastructure. The partnership l

hardwarevultr

Beyond the $7.8B in deals: Why Wall Street is suddenly watching Argentum AI

Is Argentum AI building the financing layer for the AI infrastructure boom? Yesterday, Barron’s reported that Argentum AI, the infrastructure startup led by Andrew Sobko and backed by Supermicro, had

hardwaresiliconangle

Differentiable Packing of Irregular 3D Objects with Adaptive Container Estimation

arXiv:2606.16333v2 Announce Type: replace Abstract: Most existing approaches either fix the container in advance or optimize only a single container dimension through an outer search loop, leaving the

hardwarearxiv-cs-cv

End-to-End Radar and Communication Modulation Recognition with Neuromorphic Computing

arXiv:2606.24075v1 Announce Type: cross Abstract: Although deep learning-based methods can achieve high accuracy in automatic modulation recognition (AMR) tasks, their high computational cost makes it

hardwarearxiv-cs-ai

Hardware-Oriented Inference Complexity of Kolmogorov-Arnold Networks

arXiv:2604.03345v2 Announce Type: replace Abstract: Kolmogorov-Arnold Networks (KANs) have recently emerged as a powerful architecture for various machine learning applications. However, their unique

NVIDIA and AWS Collaborate to Bring AI to Production at Scale

Building AI systems at scale is demanding, requiring low-latency inference, fast vector search, strong GPU price-performance and infrastructure that can grow without multiplying operational complexity

hardwarenvidia-blog

OpenAI and Broadcom announce chip designed for LLM inference at scale

OpenAI and Broadcom unveiled Jalapeño, OpenAI's first Intelligence Processor designed specifically for LLM inference , with early testing showing substantially better performance per watt than current

hardwarears-technica

OpenAI and Broadcom unveil Jalapeño, an LLM-optimized inference chip developed from design to manufacturing tape-out in nine months, aided by OpenAI's models (OpenAI)

OpenAI: OpenAI and Broadcom unveil Jalapeño, an LLM-optimized inference chip developed from design to manufacturing tape-out in nine months, aided by OpenAI's models — Loading... - Early testing shows

OpenAI and Broadcom unveil LLM-optimized inference chip

OpenAI and Broadcom announced a collaboration on a specialized inference chip designed to optimize the performance and efficiency of large language model deployments. The chip, referred to as 'Jalapen

hardwareopenai

OpenAI, Broadcom debut custom Jalapeño chip for AI inference

OpenAI Group PBC today revealed a custom chip called Jalapeño that it will use to power its large language models. The processor is the fruit of a collaboration with Broadcom Inc., which is no strange

hardwaresiliconangle

Ornn, which plans to launch a marketplace for GPU capacity designed to function like an exchange to trade oil contracts, raised a $33M seed led by a16z (Katherine Doherty/Bloomberg)

Katherine Doherty / Bloomberg: Ornn, which plans to launch a marketplace for GPU capacity designed to function like an exchange to trade oil contracts, raised a 33M seed led by a16z — Ornn raised 33 m

RunPod, which rents access to non-Nvidia servers, raised 100M led by Summit Partners at a 1B valuation, a source says up from $100M after its seed in 2024 (Stephanie Palazzolo/The Information)

Stephanie Palazzolo / The Information: RunPod, which rents access to non-Nvidia servers, raised 100M led by Summit Partners at a 1B valuation, a source says up from $100M after its seed in 2024 — By s

SambaNova Executive Chairman Lip-Bu Tan says the AI chip startup is set to raise 800M to 1B, sources say at a ~10B valuation, up from 2B in February (The Information)

The Information: SambaNova Executive Chairman Lip-Bu Tan says the AI chip startup is set to raise 800M to 1B, sources say at a ~10B valuation, up from 2B in February — AI server chips that effectively

TurboMPC: Fast, Scalable, and Differentiable Model Predictive Control on the GPU

arXiv:2606.24039v1 Announce Type: new Abstract: Robotics increasingly relies on GPUs for parallel simulation, large-scale learning, and neural-network inference. For model predictive control (MPC) to

Understanding the Hidden Economics of Disaggregated AI Inference

New research explores how game theory can optimize NVIDIA Dynamo disaggregated serving architectures, revealing critical performance thresholds and reducing worst-case AI inference latency by up to 7.

hardwarevultr

VoltanaLLM: Energy-Efficient and SLO-Aware Disaggregated LLM Serving via Adaptive Frequency Control and State-Space Routing

arXiv:2509.04827v3 Announce Type: replace-cross Abstract: The energy cost of Large Language Model (LLM) inference is rapidly becoming a barrier to sustainable and scalable deployment. Although modern

hardwarearxiv-cs-ai

A Differentiable Atari VCS:A Complex, Fully Known Ground Truth for Explainable AI

arXiv:2606.22447v1 Announce Type: cross Abstract: Explanation requires ground truth: to verify an account of a system we must know its inner functioning-just what is missing where explainable AI (XAI)

ACEsplat: Accelerated 3D Gaussian Scene Regression via RGB and Poses Only

arXiv:2606.22091v1 Announce Type: new Abstract: Per-scene 3D Gaussian Splatting (3DGS) enables high-fidelity rendering, but practical robotic and AR scene capture pipelines often depend on external ge

All DOGE required was contact information of the recipients to confirm that funding was not fraudulent. No validated medical funding was sto…

All DOGE required was contact information of the recipients to confirm that funding was not fraudulent. No validated medical funding was stopped. Anything that appeared to be legitimate lifesaving fun

hardwareelon-musk--x

An Analysis of Untrained Deep Reservoir Networks for Audio Surveillance

arXiv:2606.22218v1 Announce Type: cross Abstract: In this paper, we investigate untrained recurrent models from the Reservoir Computing (RC) paradigm for audio surveillance, focusing on bidirectional

An interview with Nvidia VP of Healthcare Kimberly Powell on how AI can ease doctors' workloads, help address trained medical staff shortages, and more (Cristina Criddle/Financial Times)

Cristina Criddle / Financial Times: An interview with Nvidia VP of Healthcare Kimberly Powell on how AI can ease doctors' workloads, help address trained medical staff shortages, and more — The chipma

BAC-JEPA: Label-Efficient Breast Arterial Calcification Segmentation via Synthetic Mammography-Guided Supervision

arXiv:2606.22089v1 Announce Type: new Abstract: Breast arterial calcification (BAC) on screening mammograms is an emerging cardiovascular risk biomarker, but quantitative use requires reproducible seg

hardwarearxiv-cs-cv

BatchGen: An Architecture for Scalable and Efficient Batch Inference

arXiv:2606.21712v1 Announce Type: cross Abstract: Batch inference has become a central mode of AI computation, yet existing inference engines still rely on execution models designed for interactive se

Boost Inference Performance up to 15x on NVIDIA Blackwell Using DFlash Speculative Decoding

DFlash is an open source block diffusion model for speculative decoding that significantly accelerates LLM inference on NVIDIA Blackwell GPUs by drafting entire token blocks in parallel and verifying

Build an AI Scientist for Life Science Discovery with NVIDIA BioNeMo Agent Toolkit

The NVIDIA BioNeMo Agent Toolkit enables AI scientists to use structure prediction, molecular generation, docking, sequence analysis, design, and genomics as callable tools , allowing AI agents to exe

.@cartesia runs one of the hardest inference workloads: real-time voice. Their stack has to keep long-lived streams moving, serve millions o…

.@cartesia runs one of the hardest inference workloads: real-time voice. Their stack has to keep long-lived streams moving, serve millions of audio minutes a day, and hold model latency around 90ms. T

hardwaretogether-ai--x

China’s CXMT Is Set to Challenge DRAM Incumbents

China Xin Memory Technologies (CXMT) is positioning itself as a competitor to established DRAM manufacturers through domestic production capabilities and technological development. The article likely

hardwaresemianalysis

CITADEL: CSI-Based Jamming Detection and Open-Set Classification for IIoT Networks

arXiv:2606.22939v1 Announce Type: cross Abstract: Radio frequency jamming poses a critical threat to the availability of wireless Industrial Internet of Things (IIoT) networks. Existing detection and

Concordia: JIT-Compiled Persistent-Kernel Checkpointing for Fault-Tolerant LLM Inference

arXiv:2606.23521v1 Announce Type: cross Abstract: Long-running LLM agents keep valuable state resident on GPUs: KV caches, request schedulers, communication state, and sometimes online adapters. Losin

Eli Lilly plans to use its $7.3B cash reserve to fund an 'App Store' for biotech scientists after launching a data center with 1,016 Blackwell chips in 2025 (Patrick Temple-West/Financial Times)

Patrick Temple-West / Financial Times: Eli Lilly plans to use its $7.3B cash reserve to fund an “App Store” for biotech scientists after launching a data center with 1,016 Blackwell chips in 2025 — Mo

Fast-TurboQuant: A Multiplier-Free Online Vector Quantization Approach

arXiv:2606.21448v1 Announce Type: new Abstract: As large language models scale, memory bandwidth for key-value caches and retrieval-augmented generation systems becomes a critical bottleneck. While 1-

FeLoG: Scalable and Efficient Distributed Graph Embedding with Feedback Loop Mechanism

arXiv:2606.22180v1 Announce Type: cross Abstract: Graph embedding maps graph nodes into low-dimensional vectors to support applications such as recommendation, fraud detection, and graph-based retriev

FGGS-LiDAR: Ultra-Fast, GPU-Accelerated Simulation from General 3DGS Models to LiDAR

arXiv:2509.17390v2 Announce Type: replace Abstract: While 3D Gaussian Splatting (3DGS) has emerged as a strong representation for photorealistic rendering, its vast ecosystem of assets remains difficu

FleetAgent: Teleoperation Assistant for Autonomous Fleets via Vectorized V2N Messages

arXiv:2606.21222v1 Announce Type: new Abstract: Large-scale autonomous fleets rely on teleoperation to resolve rare failures, yet streaming raw sensor data from many vehicles is costly, and remote ope

Genesis Workbench: A blueprint for industry AI in life sciences, powered by Databricks and NVIDIA

Genesis Workbench is a collaborative framework developed by Databricks and NVIDIA designed to accelerate AI adoption in the life sciences industry by providing integrated tools and best practices. The

hardwaredatabricks

GRINQH: Graded Input-based Quantization Hierarchy for Efficient LLM Generation

arXiv:2606.23419v1 Announce Type: new Abstract: Autoregressive decoding with LLMs is primarily bottlenecked by GPU memory bandwidth, especially in edge-computing settings. While quantization is essent

HERALD: High-Throughput Block Diffusion LLM Serving via CPU-GPU Cooperative KV Cache Retrieval

arXiv:2606.21633v1 Announce Type: new Abstract: Diffusion LLMs (dLLMs) improve GPU utilization over autoregressive decoding by generating multiple tokens per forward pass, but their KV cache still gro

How Telcos Build Autonomous Networks with Agentic AI

Telecommunication operators deploy autonomous agents built on telecom-domain models and running inside secure execution runtimes connected to tools, digital twins, and shared skills to understand netw

i dont think anyone is correctly doing the math around how SpaceX, the NeoCloud+NeoLab, is currently going to market? SpaceX has already rec…

i dont think anyone is correctly doing the math around how SpaceX, the NeoCloud+NeoLab, is currently going to market? SpaceX has already recouped about HALF its investment in Cursor, in compute deals.

hardwareswyx--x

LLMs write fast single-GPU kernels. Ask for a multi-GPU one and they fall apart. ParallelKernelBench (PKB) measures how they fail by benchma…

LLMs write fast single-GPU kernels. Ask for a multi-GPU one and they fall apart. ParallelKernelBench (PKB) measures how they fail by benchmarking against 87 problems pulled from real codebases includi

hardwaretogether-ai--x

Multi-cancer detection using a computationally efficient CNN with transfer learning

arXiv:2606.22400v1 Announce Type: new Abstract: This study introduces a computationally efficient convolutional neural network (CNN) architecture enhanced with transfer learning for multi-cancer detec

hardwarearxiv-cs-cv