hardware

Nvidia stock closed up 8.74% on Thursday, adding ~$440B to Nvidia's market cap, after its revenue guidance reassured investors that AI demand remains strong (CNBC)

CNBC: Nvidia stock closed up 8.74% on Thursday, adding ~$440B to Nvidia's market cap, after its revenue guidance reassured investors that AI demand remains strong — Nvidia shares rose nearly 9% Thursd

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Nvidia’s next multibillion-dollar market: Breaking the AI factory out of the data center

Nvidia Corp.’s first AI infrastructure boom concentrated enormous amounts of compute. The next opportunity will be distributing it through high-speed network fabrics. Everyone knows that I love storie

Reduce ASR inference costs by 75% with NVIDIA MPS on Amazon EC2

Serving automatic speech recognition (ASR) models at scale is costly when each request uses only a fraction of a GPU. Learn how NVIDIA CUDA Multi-Process Service (MPS) with NVIDIA Triton Inference Ser

hardwareaws-ml-blog

RoBizMarkie - lip sync test MiniMax H3

hardwarer-stablediffusion

Generated about half locally on my 5070ti. But also bought cloud GPU time to speed up overall generation. Vibecoded a rolling cutter to chop the song into chunks and then feed it in as reference audio

Scalable Multi-GPU Simulation of 3D Multicellular Growth with RNN-Based Workload Balancing

arXiv:2608.25890v1 Announce Type: cross Abstract: Detailed multicellular growth simulations based on subcellular element models (SEMs) can capture complex tissue development, but their element-level i

Sources: Nvidia paused some deals with AI cloud providers in the revenue-sharing program it announced in July; Nvidia says the program is still in place (Wall Street Journal)

Wall Street Journal: Sources: Nvidia paused some deals with AI cloud providers in the revenue-sharing program it announced in July; Nvidia says the program is still in place — The company is rethinkin

Sources: the US is probing Singapore-based Apex Logistics for its suspected role in smuggling servers with Nvidia AI chips to China; Apex says it is cooperating (Bloomberg)

Bloomberg: Sources: the US is probing Singapore-based Apex Logistics for its suspected role in smuggling servers with Nvidia AI chips to China; Apex says it is cooperating — The US government is inves

The AI storage stack gets an inference-era rethink

Artificial intelligence is changing what storage and data management platforms look like. In collaboration with Super Micro Computer Inc. and Solidigm, DataDirect Networks Inc. has introduced DDN Ente

TOPAS: Workflow-Aware Prefix-State Scheduling for Multi-Agent LLM Serving

arXiv:2608.25523v1 Announce Type: new Abstract: Prefix caching introduces a fundamental tradeoff in multi-agent large language model (LLM) serving: retaining a long system-prompt key-value (KV) cache

hardwarearxiv-cs-cl

Understanding the Energy Scaling of Large Language Model Inference Across Context Lengths and Attention Architectures

arXiv:2608.25096v1 Announce Type: new Abstract: The growing adoption of large language models (LLMs) has raised increasing concerns about the energy consumption and environmental impact of inference.

A Feature-Major Codebook for Memory-Efficient Sparse-Binary Self-Organizing Maps: Scaling a MEDLINE Atlas to 1.05 Million Neurons on a Single Consumer GPU

arXiv:2608.24067v1 Announce Type: new Abstract: A self-organising map turns a large corpus into a browsable two-dimensional atlas, but building one at MEDLINE scale has been impractical: the best-matc

Amazing! 🥳 Thanks @vllm_project for getting Qwen3.8-Flash-Next running on NVIDIA and AMD from day 0.

Amazing! 🥳 Thanks @vllm_project for getting Qwen3.8-Flash-Next running on NVIDIA and AMD from day 0. Qwen3.8-Flash-Next from @Alibaba_Qwen has day-0 support in vLLM, verified on NVIDIA and AMD GPUs. 🎉

hardwareqwen--x

Bayes with No Shame: Admissibility Geometries of Predictive Inference

arXiv:2603.05335v3 Announce Type: replace-cross Abstract: Modern predictive systems combine predictors, sequential monitors, prediction sets, and online strategies, each with a different certificate o

Bring your own model with Amazon SageMaker AI: Script mode in SDK v3

The SageMaker Python SDK v3 redesigns script mode with unified ModelTrainer and ModelBuilder classes. This post walks through two end-to-end examples, a scikit-learn Random Forest and a multi-GPU Stab

hardwareaws-ml-blog

Deep Feature Pyramid Convolutional Networks with In-Place Activated Batch Normalization for Automated Skin Lesion Boundary Segmentation

arXiv:1812.00877v2 Announce Type: replace Abstract: Segmentation of skin lesion boundaries in dermoscopic imaging is an important prerequisite step for computer-aided diagnosis of malignant melanoma,

Elastic KV Cache for LLM Serving:A Working Reclamation Mechanism, and Why Chunked Prefill Already Closes the Gap

arXiv:2608.23658v1 Announce Type: cross Abstract: An LLM serving engine sizes its key-value (KV) cache once, at startup, permanently setting aside a reserve for the worst-case prefill activation. Duri

Indian AI infrastructure company AM Intelligence orders 9,000 Nvidia Vera Rubin systems and plans to offer 1GW of computing capacity as part of an $8B project (Saritha Rai/Bloomberg)

Saritha Rai / Bloomberg: Indian AI infrastructure company AM Intelligence orders 9,000 Nvidia Vera Rubin systems and plans to offer 1GW of computing capacity as part of an $8B project — AM Intelligenc

Introducing v5 Droplets: next-generation performance, sized to your workload

We’re excited to introduce v5 Droplets, a new generation of compute built on 5th Gen AMD EPYC™ processors. v5 Droplets are purpose built to deliver higher performance for demanding workloads such as c

hardwaredigitalocean

Learning the Kohn-Sham map with neural operators for quasi-linear scaling density functional theory

arXiv:2608.23895v1 Announce Type: cross Abstract: Kohn--Sham density functional theory (DFT) underpins electronic-structure simulations, but repeated orbital diagonalizations lead to cubic scaling, re

Minima-KV: Retention-Preserving KV Cache Compression with Mixed-Format Paged Attention

arXiv:2608.23834v1 Announce Type: new Abstract: The key-value (KV) cache is a primary capacity and bandwidth bottleneck in long-context LLM serving. We present Minima-KV, a retention-preserving hierar

Nvidia doubles its revenue as demand for AI chips accelerate, but bubble fears persist

Nvidia Corp. reported better-than-expected second-quarter earnings and revenue and also delivered bullish guidance for the current quarter and next full year, sending its stock up more than 4% in afte

Nvidia forecasts Q3 revenue of 108B, above a 104B estimate, and expects FY 2028 revenue to grow by about 70%; NVDA jumps 4%+ after hours (Deborah Mary Sophia/Reuters)

Deborah Mary Sophia / Reuters: Nvidia forecasts Q3 revenue of 108B, above a 104B estimate, and expects FY 2028 revenue to grow by about 70%; NVDA jumps 4%+ after hours — Nvidia's (NVDA.O) quarterly re

NVIDIA NVLink Fusion Brings NVHBM to Next-Generation AI Infrastructure

NVIDIA’s NVLink Fusion integrates its custom NVHBM memory into the MGX rack‑scale architecture, enabling hyperscalers to deploy large numbers of custom XPUs with higher bandwidth, reduced silicon area

hardwarenvidia-developer

Output-Level Regularization Eliminates the Seed Lottery in Single-GPU VLA Fine-Tuning

arXiv:2606.13856v2 Announce Type: replace Abstract: Fine-tuning a vision-language-action model (VLA-JEPA) on a single GPU should be simple: load a pretrained checkpoint, run training, deploy. There is

hardwarearxiv-cs-ro

Qwen3.8 27B C8 at 972 TG / 5,680 PP on 4x MI100 rig ($6.5k) using my new INT8 vLLM fork

Yet another vLLM fork thread here, but this time its for older INT8-centric hardware. This is a complete INT8 serving stack for Qwen3.8 27B based on vLLM, AITER, and a 27B GPTQ INT8 quant w/ DFlash2.

hardwarer-localllama

Rethinking Pre-Training and Augmentation for Zero-Shot Cross-City Object Detection

arXiv:2608.24154v1 Announce Type: cross Abstract: Real-world deployment of traffic surveillance systems is bottlenecked by geographic domain shift, in which models trained in one city underperform whe

RT-NeuS: Towards Real-Time Neuro-Symbolic Video Understanding via Adaptive Temporal Verification

arXiv:2602.23553v2 Announce Type: replace Abstract: Long-form video question answering (LVQA) requires answering natural-language queries about videos spanning minutes to hours, demanding temporal rea

SatDL: Jointly Optimizing Data Redistribution and Training for Satellite-Based Distributed Learning

arXiv:2608.24516v1 Announce Type: cross Abstract: Satellite-based distributed learning promises to train machine-learning models directly in orbit using massive, globally dispersed sensor data, thereb

Scalable Question-Centric Text-to-Image Evaluation: Reliable Ranking, Fine-Grained Diagnosis, and Cost-Aware Routing

arXiv:2608.24112v1 Announce Type: new Abstract: Modern text-to-image (T2I) models often have similar total scores but different strengths, making practical selection difficult. Fine-grained benchmarks

Serving Masked Diffusion LLMs: Characterization and Design Principles from Real Hardware

arXiv:2608.23807v1 Announce Type: new Abstract: Masked diffusion language models (dLLMs) can in principle generate text faster than autoregressive (AR) models, since they denoise many tokens at once.

ShardMeter: Sharded and Geo-Distributed Training Without the Guesswork

arXiv:2608.23840v1 Announce Type: cross Abstract: Training large-scale AI models often outgrows a single data center, demanding sharded, multi-cluster, and decentralized training. However, the huge sp

Source: Nvidia has agreed to buy Hugging Face for $12.9B; the AI repository has had several potential suitors among its investors, including Salesforce (The Information)

The Information: Source: Nvidia has agreed to buy Hugging Face for 12.9B; the AI repository has had several potential suitors among its investors, including Salesforce — Nvidia has agreed to buy Huggi

Source: Nvidia has been in talks to acquire Hugging Face for more than $13B; Microsoft also met with Hugging Face, but talks are not ongoing (Business Insider)

Business Insider: Source: Nvidia has been in talks to acquire Hugging Face for more than $13B; Microsoft also met with Hugging Face, but talks are not ongoing — - The companies have not reached a deal

Sources: Anthropic has agreed to pay Nscale $45B over six years to rent about 460MW of power at a West Virginia data center using Nvidia's Vera Rubin chips (Brody Ford/Bloomberg)

Brody Ford / Bloomberg: Sources: Anthropic has agreed to pay Nscale 45B over six years to rent about 460MW of power at a West Virginia data center using Nvidia's Vera Rubin chips — Anthropic PBC has a

The network becomes the computer as Cisco and Nvidia accelerate AI factory rack-scale

The first phase of the generative artificial intelligence infrastructure boom was defined by a race for graphics processing units. The next phase will be defined by what happens after those GPUs arriv

A detailed look at Jalapeño, OpenAI's ASIC developed with Broadcom in 16 months, which beat Nvidia, AMD, and Google chips on multiple top open-weight models (SemiAnalysis)

SemiAnalysis: A detailed look at Jalapeño, OpenAI's ASIC developed with Broadcom in 16 months, which beat Nvidia, AMD, and Google chips on multiple top open-weight models — OpenAI's self-designed ASIC

Adversarial Entropy Inflation Against Gumbel-Based Inference Verification

arXiv:2608.23375v1 Announce Type: cross Abstract: Gumbel-based inference verification bounds LLM weight exfiltration by only forgiving token choices that plausibly arise from honest GPU nondeterminism

AI factories enter the execution era as Cisco and Nvidia push rack-scale systems into production

The artificial intelligence infrastructure market is crossing an important threshold. The conversation is shifting from acquiring graphics processing units to building complete AI factories that can g

BIMScript: Material-Aware Structured Scene Programs for BIM Ingestion

arXiv:2608.21447v1 Announce Type: new Abstract: Structured-language models such as SceneScript reconstruct a scene as a short program of parametric commands, an inherently editable and semantically ex

Cisco and Nvidia take AI factories from rack to runtime

AI factories are moving from ambitious plans toward production, but the path from graphics processing unit acquisition to usable systems remains a race against time. Neoclouds already have customers w

Cisco expands rack-scale secure AI factory infrastructure for neocloud and sovereign clouds

Cisco Systems Inc. today introduced an expansion to its current lineup of rack-scale artificial intelligence solutions for massive AI workloads in secure AI factories, addressing growing needs for mas

CST: Collaborative Selective Transmission for Communication-Efficient Multimodal Edge Inference

arXiv:2608.22115v1 Announce Type: new Abstract: Collaborative multimodal inference improves edge perception by combining observations from distributed sensing devices, but transmitting high-dimensiona

CUDA Python 1.0: Stable APIs, One Foundation, Full Platform Access

CUDA Python 1.0, released alongside CUDA 13.3, provides NVIDIA-maintained libraries and tools that expose the entire CUDA platform directly to Python through a single shared foundation. It adopts sema

hardwarenvidia-developer

Democratizing the AI data center: How Cisco and Nvidia are bringing rack-scale power to the enterprise

For the past two years, the generative artificial intelligence boom has largely been defined by a single market. Hyperscalers have dominated the market with massive capital spending. According to Cred

FlashReg: GPU-Accelerated 3-Clique Point Cloud Registration for Real-Time Correspondence-to-Pose Estimation

arXiv:2608.21804v1 Announce Type: new Abstract: Graph-based point cloud registration achieves high robustness by identifying geometrically consistent correspondence sets, but constructing second-order

Had a lot of fun chatting again with my twin brother @dylan522p We went through lab economics over the next few years - the shift from infer…

Had a lot of fun chatting again with my twin brother @dylan522p We went through lab economics over the next few years - the shift from inference to training as RSI draws near; and how Anthropic and Op

hardwaredylan-patel--x

If you're looking for a weekend project, how about training your own text-to-speech model from scratch on your own GPU, and then running it …

If you're looking for a weekend project, how about training your own text-to-speech model from scratch on your own GPU, and then running it on any device's CPU? We just open-sourced the entire Pocket

hardwareyann-lecun--x

Kernel Token Contradiction: a Fast and Principled Approach for LLM Claim Uncertainty Quantification

arXiv:2608.22506v1 Announce Type: new Abstract: Claim-level Uncertainty Quantification (UQ) aims to mitigate the lack of reliability of Large Language Models (LLMs) by evaluating the factuality of eac

hardwarearxiv-cs-cl