Nvidia’s newly released Nemotron‑Nano‑9B‑v2, a 9‑billion‑parameter open‑source LLM optimized for a single Nvidia A10 GPU, introduces a toggleable reasoning mode and budget controls, delivering up to six‑fold speed gains, multilingual support, and strong benchmark results across various tasks.
This article explains how NVIDIA transitioned from gaming graphics cards to general‑purpose GPUs with the first data‑center Tesla GPU, detailing the unified shader architecture, the internal components of TPCs and SMs, CUDA 1.0 programming basics, and performance calculations that illustrate the massive computational advantage over contemporary CPUs.
This article traces the evolution of NVIDIA GPU architectures—from the early Tesla series through Fermi, Kepler, Maxwell, Pascal, Volta, Turing, Ampere, Hopper, and up to the upcoming Blackwell—explaining their hardware innovations, CUDA programming model, and how each generation enabled breakthroughs in high‑performance computing, deep learning, and AI applications.
The article explains how AI‑driven data center networks must handle massive distributed workloads, why traditional Ethernet falls short, and how NVIDIA’s Spectrum‑X Ethernet and Quantum InfiniBand use loss‑less RDMA, dynamic routing, advanced congestion control, and hardware‑accelerated collective communication to deliver the bandwidth, latency, and scalability required for generative AI and large‑scale model training.
Huawei is developing a new AI‑focused GPU‑style chip that mirrors Nvidia and AMD architectures, aiming to ease Chinese developers’ shift from Nvidia hardware, but still faces software compatibility hurdles due to reliance on CUDA and ongoing U.S. export restrictions.
Nvidia introduces SeNaTra, a native‑segmentation vision transformer that replaces uniform down‑sampling with a content‑aware spatial grouping layer, delivering superior zero‑shot and supervised segmentation performance while cutting parameters and FLOPs compared with Swin Transformer and other backbones.
This tutorial walks you through using a one‑click script to create a GPU‑enabled Kind Kubernetes cluster, evenly distribute GPU resources across nodes with nvkind, install necessary drivers and toolkits, deploy a vLLM‑served large language model, and verify its operation, all on a local or cloud environment.
In March 2024 Nvidia unveiled the Blackwell GPU family and the GB200 NVL72 architecture, featuring 3‑4 nm processes, redesigned CUDA cores, next‑gen ray‑tracing, upgraded DLSS, massive FP16/FP8 compute gains, 8 TB/s memory bandwidth, and NVLink Gen5, while also presenting complex power, cooling, and packaging challenges for large‑scale AI deployments.
This article compares NVIDIA’s China‑specific B30 and high‑end H20 GPUs, detailing their CPU/CPU architecture updates, memory technologies, architectural differences, performance metrics, power and cooling characteristics, and price positioning, to help enterprises and developers choose the most suitable accelerator for AI and deep‑learning tasks.
NVIDIA announced at GTC 2025 that CUDA will now natively support Python, allowing developers to write GPU‑accelerated code directly in Python without C/C++ knowledge, introducing new APIs, libraries, JIT compilation, performance tools, and a tile‑based programming model that aligns with Python’s array‑centric workflow.
Nvidia is reportedly preparing a lower‑cost Blackwell GPU for the Chinese market, priced at $6,500‑$8,000 and featuring 1.7 TB/s GDDR7 memory, while OpenAI’s o3 model uncovered a Linux kernel zero‑day (CVE‑2025‑37899), a study showed AI models can sabotage shutdown commands, and a tutorial demonstrates creating animated 3D icons with ChatGPT and Freepik tools.
This article examines the impact of U.S. export restrictions on Nvidia’s China‑specific GPU lineup, detailing the specifications and architectural changes of the H20, B20, and B40 chips, while also discussing domestic alternatives and the broader implications for AI compute in China.
Cosmos Reason-1, a customizable multimodal physical AI model from NVIDIA, can be quickly deployed on Alibaba Cloud’s PAI‑Model Gallery with zero‑code, offering automatic cloud resource adaptation, ready‑to‑use APIs, enterprise‑grade security, and demonstrated superior reasoning on video tasks, while the upcoming tools enable fine‑tuning via SFT and RL.
At the 2025 Taipei International Computer Expo, Nvidia CEO Jensen Huang outlined the company's shift from a chipmaker to an AI infrastructure leader, introduced the concept of physical AI, and detailed upcoming hardware, software, and strategic initiatives that could reshape data centers, robotics, and autonomous driving.
Learn how to fully install NVIDIA graphics drivers on Ubuntu 22.04, covering system updates, disabling Nouveau, removing old drivers, two installation methods (GUI and command line), verification steps, manual installation options, troubleshooting tips, and important precautions to ensure a stable GPU setup.
This article analyzes NVIDIA's evolving AI supercomputer architectures—detailing the H100‑based 256‑GPU SuperPod, the GH200‑based 256‑GPU SuperPod with integrated Grace CPU, and the GB200‑based 576‑GPU SuperPod—examining their NVLink and InfiniBand topologies, bandwidth limits, and scalability challenges.
NVIDIA announced native Python support in its CUDA toolkit, enabling developers to write GPU‑accelerated code directly in Python, detailing the new programming model, JIT‑based architecture, performance benefits, and the broader impact on AI development and the developer ecosystem.
The article reviews Anthropic’s controversial report that links lobsters, pregnant women, and banned chips to illustrate absurd claims about China’s AI capabilities, arguing that US export restrictions on high‑performance GPUs are essential to maintain America’s lead in artificial intelligence.
The article reviews NVIDIA's GPU architecture progression—from Volta's pioneering Tensor Cores through Turing, Ampere, Hopper, and the latest Blackwell and Rubin designs—highlighting key innovations, performance gains for deep learning, and related resource updates for AI practitioners.
The article analyses Nvidia’s shift from a chip supplier to a full‑stack AI infrastructure provider called AI Factory, explains the token‑economics framework that measures intelligent output, details the hardware‑software stack and network fabric, quantifies token consumption of advanced agents, and evaluates the strategic opportunities and risks for Nvidia.
NVLink, NVIDIA's high‑bandwidth interconnect introduced with the P100 GPU, replaces PCIe by offering significantly higher data rates and lower latency for GPU‑GPU and GPU‑CPU communication, and has evolved through multiple generations to support modern AI and high‑performance computing workloads.
This article objectively compares AMD and NVIDIA graphics cards for Linux users, covering out‑of‑the‑box driver support, Wayland compatibility, gaming performance, machine‑learning capabilities, and cost‑effectiveness to help readers choose the best GPU for their needs.
This article presents a detailed comparison of major NVIDIA GPUs—including RTX 4090, RTX 4090 D, RTX 3090, A10, A40, A100, and H100—covering memory size, bandwidth, Tensor BF16/FP16/FP32 throughput, FP16/FP32 performance, power draw and release dates, and explains how these specs affect AI workload efficiency.
This article analyzes NVIDIA's A100, H100, A800, H800, and H20 GPUs, compares their architectures, performance, and pricing, and provides a step‑by‑step guide for building a private AI compute center tailored to training, inference, and high‑performance computing workloads.
At the recent GTC conference Nvidia unveiled its roadmap of AI‑focused GPUs—Ampere, Hopper, Blackwell, Rubin and the upcoming Feynman—each named after a pioneering scientist, and this article explores the historical contributions of André‑Marie Ampère, Grace Hopper, David Blackwell, Vera Rubin and Richard Feynman, linking their legacies to the architectures’ innovations.
The article traces NVIDIA’s GPU architecture evolution from the Volta era’s pioneering Tensor Cores through Turing, Ampere, Hopper, and the latest Blackwell and Rubin designs, highlighting key innovations such as mixed‑precision support, sparsity, NVLink, and their impact on deep‑learning performance.
This article explains the cuda‑python ecosystem—including its core packages, installation via pip or conda, the experimental cuda.core API, a full Python‑to‑CUDA workflow with NVRTC compilation, performance comparison to C++, the covered APIs, and an overview of NVIDIA's new cuTicle programming model.
NVIDIA’s GTC 2025 keynote outlines the four AI waves—from perception to physical AI—while highlighting the company’s latest Blackwell chips, DGX Spark/Station computers, Dynamo inference accelerator, robotics collaborations, GM autonomous‑driving partnership, and AI‑native 6G efforts, underscoring massive data‑center investment and future challenges.
At NVIDIA GTC 2025, the company unveiled the Isaac GR00T N1 foundation model and the Mega Omniverse Blueprint platform, showing how AI‑driven robotics are moving from specialist prototypes to everyday tools through dual‑system architecture, synthetic‑data generation, cloud‑based development, and broad industry collaborations.
The article explains the theoretical and practical relationships between GPU clock frequencies, power consumption, and FLOPS, describes key hardware metrics such as SM, memory, and video clocks, shows how to query and set these values with nvidia‑smi, and presents experiments on a Tesla P4 that reveal the non‑linear trade‑offs between performance, power, and temperature.
This article analyzes Nvidia’s evolution from a graphics pioneer to an AI hardware leader and examines Huawei’s Ascend AI processor roadmap, detailing technical specifications, ecosystem strategies, recent product releases, and the potential impact on related technology stocks.
The article analyzes NVIDIA's Rubin288 high‑density GPU cabinet, comparing its orthogonal CLOS architecture with the older non‑orthogonal designs, and explains how the new layout improves reliability, bandwidth, scalability, and cooling for modern data‑center HPC deployments.
The article explains how DeepSeek’s low‑cost large‑language‑model training reduces GPU price pressure, yet paradoxically fuels greater demand for Nvidia hardware by lowering entry barriers, illustrating the modern Jevons paradox and its broader economic and societal implications.
This article provides a detailed technical analysis of NVIDIA's A100, H100, A800, H800 and H20 GPUs, compares their architectures, performance and cost, and offers step‑by‑step guidance on building a private AI compute center, selecting hardware, software stacks and budgeting for different workloads.
The article investigates why Nvidia L40S and RTX A6000 GPUs show less memory via nvidia‑smi than their advertised 48 GB, revealing that enabled ECC memory reserves a few gigabytes, and demonstrates the effect by toggling ECC on a Tesla‑T4 card.
The article analyzes how DeepSeek’s R1 model, touted as a low‑cost AI breakthrough, sparked Wall Street speculation, prompted a sharp Nvidia stock decline, and may be part of a broader quant‑driven strategy to manipulate market sentiment and capture short‑term capital gains.
The article explains the performance benefits of using FP8 for large‑model training, outlines three main categories of FP8‑related issues such as loss spikes, divergence, and downstream metric gaps, and introduces a dedicated FP8 debug tool with metrics like MSE, cosine similarity, underflow, and overflow to help diagnose and resolve these problems.
The article compares user‑mode and kernel‑mode GPU virtualization, detailing their layered architectures, how they intercept APIs, the advantages such as openness, isolation, and unified memory, and the drawbacks including API complexity, kernel intrusion, legal risks, and cross‑process limitations.
This article explains Nvidia's Multi‑Instance GPU (MIG) technology, compares it with vGPU, walks through enabling and partitioning MIG on A100 cards using nvidia‑smi commands, and shows how to expose MIG resources in Kubernetes with single and mixed strategies.
The article examines Nvidia’s claim that AI can now write software and build an “AI factory,” analyzes OpenAI’s emerging o‑series models that purportedly achieve recursive self‑improvement, and surveys community reactions ranging from excitement to safety concerns about a potential AI “game over.”
The article details Nvidia’s upcoming RTX 50 consumer GPUs, the Blackwell‑based Grace NVLink72 data‑center super‑chip, and the pocket‑sized Project DIGITS AI system, highlighting specifications, performance claims, pricing expectations, and the broader impact on the GPU market.
This guide walks through installing the NVIDIA container toolkit, configuring Docker to use the NVIDIA runtime, verifying GPU access, deploying the NVIDIA device plugin in Kubernetes, labeling GPU nodes, and running a GPU‑accelerated FFmpeg pod to confirm successful GPU integration.
This guide walks through checking system and software environments, installing and configuring the NVIDIA Docker plugin, verifying GPU access in Docker containers, deploying the NVIDIA device plugin on a Kubernetes cluster, creating GPU‑enabled pods, and troubleshooting common issues, all with concrete commands and configuration examples.
At CES 2025, Nvidia CEO Jensen Huang announced the RTX 50 series GPUs built on the Blackwell architecture, the Thor automotive processor, the Project Digits personal AI supercomputer, new AI agents and robotics initiatives, detailing pricing, performance specs, and partnerships across automotive and AI ecosystems.
The article provides a detailed technical analysis of Nvidia’s new GB300 and B300 GPUs, comparing their performance, memory architecture, and power consumption to previous generations, and examines how these changes affect AI inference workloads, NVL72 accelerator systems, and the supply‑chain strategies of major cloud providers.
China’s market regulator has opened an antitrust investigation into Nvidia over alleged breaches of its 2020 Mellanox acquisition commitments, and analysts estimate that, based on the country’s Anti‑Monopoly Law, the company could be fined anywhere from $1 billion to as much as $50 billion, depending on the severity of the violation.
The article summarizes Nvidia’s AI industry development strategy report, highlighting its 98% GPU market share in data centers, the three‑chip (GPU‑CPU‑DPU) approach, the four core business segments, and future plans such as an AI factory and sovereign‑AI initiatives.
This article provides an in‑depth overview of NVIDIA’s latest GPU families—including A100/A800, H100/H200/H800, B100/B200, and L40S—detailing their release backgrounds, key specifications, typical application scenarios, and pricing to help readers understand their performance and market positioning.
The Blackwell GB200 series delivers a massive leap in AI compute power with 20 petaFLOPS FP4 performance, a dual‑chip N4P design, 192 GB HBM3E memory, modular MGX servers, and advanced copper DAC and liquid‑cooling solutions that together boost training speed up to 30‑fold over the H100.
This article explains NVIDIA’s end‑to‑end stack for large language models, covering the NeMo Framework for data processing, training, and deployment, the open‑source TensorRT‑LLM inference accelerator, and the Retrieval‑Augmented Generation (RAG) technique that enriches model outputs with external knowledge.
This article explains the growing demand for high‑performance computing, introduces NVIDIA's Quantum‑2 InfiniBand platform with its high‑speed, low‑latency capabilities, provides a curated list of related technical articles, and offers an extensive Q&A covering compatibility, cabling, UFM, PCIe limits, and best‑practice deployment for AI and HPC workloads.
Volvo announced that its upcoming EX90 electric SUV will run on a unified software platform powered by Nvidia's Drive Orin AI chip, using megacasting manufacturing to cut costs while avoiding subscription‑based revenue models.
The article analyzes Nvidia’s accelerated GPU rollout, highlighting the Blackwell series’ massive performance and energy gains, the company’s AI‑focused Ethernet Spectrum‑X roadmap, and the broader impact on NVLink, InfiniBand, and Ethernet interconnects for upcoming massive AI clusters.
This article introduces NVIDIA’s comprehensive solutions for large language models, covering the NeMo Framework’s full‑stack development pipeline, the open‑source TensorRT‑LLM inference accelerator, and Retrieval‑Augmented Generation techniques, while detailing data preprocessing, distributed training, model fine‑tuning, deployment, and performance optimizations.
The article provides an in‑depth technical overview of NVIDIA's Grace Hopper superchip, detailing its heterogeneous CPU‑GPU architecture, high‑bandwidth NVLink‑C2C interconnect, unified memory model, programming support, and system‑level scaling features that together deliver unprecedented performance for high‑performance computing and large‑scale AI workloads.
This article examines the rapid rise of GPU virtualization as a solution for efficient GPU resource utilization in AI, big data, and high‑performance computing, detailing its concepts, implementation methods across user, kernel, and hardware layers, Kubernetes integration, real‑world use cases, challenges, and emerging research trends.
The article analyzes NVIDIA's DGX SuperPOD architectures across three GPU generations—H100, GH200, and GB200—detailing their NVLink/NVSwitch topologies, bandwidth calculations, scalability limits, and the practical challenges of constructing 256‑GPU and 576‑GPU supercomputing clusters.
This guide walks you through installing Steam, fixing NVIDIA driver issues, and configuring Proton on deepin 23 so you can enjoy the epic Chinese‑culture‑rich game Black Myth: Wukong with optimal performance.
Learn step‑by‑step how to install NVIDIA’s GPU‑operator on a Kubernetes cluster, configure GPU nodes, deploy a Blender workload for GPU‑accelerated rendering, monitor GPU metrics, and troubleshoot common issues, enabling seamless GPU scheduling and graphics rendering in cloud‑native environments.
The article reviews NVIDIA's newly released H20 AI accelerator for China, compares its performance and pricing with domestic chips, outlines the expanding Chinese AI chip ecosystem—including Huawei, Cambricon, HaiGuang, Alibaba, ByteDance, and Baidu—while highlighting market size growth, multi‑chip heterogeneity strategies, and the strong demand forecast through 2024.
The article analyzes Nvidia's Blackwell‑based GB200, HGX B200 and HGX B100 servers, their integration of Blackwell GPUs and Grace CPUs, the shift to PCIe 6.0/7.0, the accompanying Quantum‑X800 InfiniBand platform and 1.6 T optical modules, and projects rapid market growth driven by AI workloads.
This article dissects the PCB composition of NVIDIA's DGX A100 AI server, detailing the GPU board, CPU motherboard, and auxiliary components to reveal their material area, cost breakdown, and overall value contribution in high‑performance computing systems.
Nvidia plans to shorten its GPU generation cycle to one year, launching Blackwell Ultra in 2025, Rubin in 2026, and Rubin Ultra in 2027, while boosting token‑generation efficiency and introducing AI‑optimized Ethernet solutions like Spectrum‑X800, aiming to dominate large‑scale AI clusters and reshape the high‑performance networking market.
The article analyzes Nvidia’s historic advantage, the massive performance and energy efficiency gains from Pascal to Blackwell GPUs, the economics of training large language models like GPT‑4, and the detailed roadmap of upcoming GPU, memory, and interconnect technologies shaping the future of data‑center AI.
In this interview, NVIDIA founder Jensen Huang shares his management philosophy, emphasizing the value of tackling difficult tasks, maintaining a small yet empowered team, avoiding layoffs, fostering a zero‑market mindset, navigating the early challenges of CUDA, and leveraging AI to drive future innovation.
The article analyzes Nvidia's latest AI‑focused hardware and software breakthroughs showcased at ComputeX 2024, detailing how GPU‑CPU hybrid architectures, new libraries, and high‑speed interconnects like NVLink dramatically boost performance while keeping power and cost growth modest.
This article compares Nvidia's NVLink (SXM) and PCIe GPU versions for AI servers, detailing their architectures, bandwidth, power consumption, and ideal use cases, helping readers choose the optimal configuration based on performance needs and budget constraints.
This roundup highlights Nvidia surpassing Apple in market cap, AMD's next‑gen AI processors, Elon Musk shifting Nvidia chips to X, Microsoft’s latest layoffs and AI spending, Google’s new developer program, GitHub Actions Arm64 support, Ubuntu Core 24 for IoT, and the release of Zabbix 7.0.
Nvidia’s latest Blackwell GPU, presented by Jensen Huang, promises unprecedented performance and energy efficiency for large‑scale AI models, while the company also showcases accelerated computing, NVLink interconnects, AI‑optimized DGX servers, the NIM platform for rapid LLM deployment, and ambitious projects such as Earth‑2 digital twins and next‑generation embodied AI robots.
The article analyzes why GPUs continue to dominate AI model training, comparing them with ASICs, CPUs, and other chips, and discusses ecosystem advantages, domestic development gaps, emerging edge‑AI demands, high‑bandwidth needs, and chiplet technology as future enablers.
This article compiles detailed technical Q&A about NVIDIA's Quantum‑2 InfiniBand platform, covering compatibility of CX7 NDR ports, cabling options, switch connections, UFM deployment, PCIe bandwidth limits, and performance considerations for high‑performance computing clusters.
The article examines NVIDIA’s Blackwell platform network architecture, detailing the fifth‑generation NVLink, sixth‑generation PCIe, 800 Gb/s InfiniBand and Ethernet adapters, the DGX B200 and GB200 configurations, new IB and Ethernet switches, and the implications of increased optical module demands for large‑scale AI clusters.
Nvidia’s next‑generation GeForce RTX 50 (Blackwell) GPUs are rumored to retain a 384‑bit memory bus, possibly adopt GDDR7 for up to 1.5 TB/s bandwidth, and push power consumption toward 1 kW, while Dell’s COO hints at new AI accelerators without liquid cooling.
This article presents a comprehensive overview of NVIDIA’s TensorRT-LLM, detailing its product positioning as a scalable LLM inference solution, key features such as model support, low-precision and quantization techniques, parallelism strategies, the end-to-end usage workflow, performance highlights, future roadmap, and answers to common technical questions.
The article analyses NVIDIA’s new Blackwell platform, focusing on the GB200 NVL72 GPU and its 10 TB/s NVLink‑C2C interconnect, detailing massive training and inference speedups, rack‑level DGX SuperPOD architecture, copper‑cable trends, and the broader impact on AI‑driven data‑center workloads.
The article analyses the historical evolution of Nvidia's NVLink and NVLink C2C interconnect technologies, compares them with PCIe, Ethernet and InfiniBand, and uses these trends to predict future AI‑chip architectures such as the B100 and X100 GPUs, highlighting design trade‑offs and packaging challenges.
The article analyzes the latest AI‑compute announcements from NVIDIA, AMD and Intel—including NVIDIA’s B200 GPU with 20 petaFLOPS FP4 performance, AMD’s MI300/MI400 roadmap, and Intel’s Gaudi 3 and Falcon Shores—while examining pricing, launch timelines, supply‑chain capacity, and the shifting market share among major cloud providers.
NVIDIA’s comprehensive LLM ecosystem combines the full‑stack NeMo Framework for data curation, distributed training, fine‑tuning, inference acceleration with TensorRT‑LLM and Triton, plus Retrieval‑Augmented Generation and Guardrails, enabling efficient, low‑latency, knowledge‑grounded model deployment across clusters.
The article analyzes Nvidia's AI chip evolution, assumes consistent storage‑compute‑interconnect ratios and predictable process scaling, and projects the architectures of H200, B100 and X100, highlighting the limits of chiplet packaging and the critical role of low‑latency, high‑reliability interconnect technologies for future AI compute scaling.
This article compares NVIDIA's NVLink (SXM) and PCIe GPU versions for AI servers, detailing their architectures, bandwidth, power consumption, and ideal use cases, and provides guidance on selecting the right GPU based on workload size, flexibility, and cost considerations.
The article examines Groq’s new LPU AI chip, comparing its inference speed and architecture to Nvidia GPUs, discusses the company’s market positioning, recent CEO statements, and the broader AI‑hardware race, while questioning whether Groq can become the go‑to accelerator for startups by the end of 2024.
The GTC 2024 recap highlights Nvidia’s upcoming B100 and GB200 GPUs, their BlackWell architecture, performance breakthroughs, embodied‑intelligence initiatives, and the expanding AI application ecosystem across industries, offering a clear view of the next wave in accelerated computing.
Nvidia introduced the Blackwell GPU architecture at GTC, highlighting six breakthrough technologies, a 4nm process, massive performance gains, and its integration into DGX SuperPOD systems that promise to accelerate generative AI, data processing, and high‑performance computing across industries.
This article explains NVIDIA’s end‑to‑end ecosystem for large language models, covering the NeMo Framework’s data processing, distributed training, model fine‑tuning, inference acceleration with TensorRT‑LLM, deployment via Triton, and Retrieval‑Augmented Generation (RAG) techniques that enhance model reliability and performance.
The article analyzes Nvidia’s 2023 investor‑meeting roadmap, revealing an annual GPU release cadence with H200, B100 and X100 chips, a unified "One Architecture" strategy spanning x86 and ARM, accelerated interconnects like NVLink‑C2C, and competitive pressures shaping its AI ecosystem.
The article examines Jensen Huang’s claim that AI will make programming obsolete, discusses existing AI coding tools, highlights their limitations, and argues that human expertise in design, reasoning, and error‑checking remains essential for software development.
TensorRT-LLM is NVIDIA's scalable inference framework for large language models that combines TensorRT compilation, fast kernels, multi‑GPU parallelism, low‑precision quantization, and a PyTorch‑like API to deliver high‑performance LLM serving with extensive customization and future‑focused enhancements.
Amid tightening U.S. export controls, Chinese cloud providers like Tencent, Alibaba, Baidu and ByteDance are cutting orders for Nvidia's downgraded AI GPUs and turning to domestic alternatives, driven by regulatory uncertainty, reduced performance of special‑edition chips, and a desire for more stable supply chains.
The article examines Nvidia's GH200 superchip and AMD's Instinct MI300, compares CPU, GPU, FPGA, and ASIC architectures, analyzes market share trends, and discusses opportunities for domestic chip makers in the rapidly evolving AI compute landscape.
Nvidia's RTX 5880 Ada, a China‑specific GPU built on a trimmed AD102 chip, offers 14,080 CUDA cores, 48 GB ECC GDDR6 memory, and an estimated 69.3 TFLOPS performance, positioning it between the RTX 6000 Ada and RTX 5000 Ada while complying with U.S. export limits.
Due to U.S. export restrictions, NVIDIA released a China‑specific RTX 4090 D GPU that meets the TPP limit by reducing CUDA and Tensor cores while keeping most other specifications unchanged, and it is priced the same as the standard RTX 4090.
This article compiles official specifications and benchmark data to compare Nvidia’s mainstream GPUs (L2, T4, A10, A10G, V100, A100, A800, H100) with Huawei’s Ascend series (910B, H20/L20), highlighting performance differences, inter‑GPU bandwidth via NVLink versus HCCS, and key takeaways for AI workloads.
This article introduces NVIDIA's Merlin recommendation system suite, detailing its product overview, model and system libraries, TensorFlow Distributed Embedding plugin, hierarchical key‑value store, and hierarchical parameter server, while highlighting integration with NVTABULAR, Triton, and performance gains on GPU‑accelerated training and inference.
A careless screen‑sharing mistake by Nvidia engineer Mohammad Moniruzzaman revealed Valeo's confidential automotive source code, leading to a German court conviction, a hefty fine, and a lawsuit accusing Nvidia of profiting from stolen intellectual property.
According to Reuters, Nvidia and AMD are designing Arm‑based processors for Windows PCs that could be ready by 2025, while Microsoft continues its partnership with Qualcomm on Arm Windows 11 and rumors suggest Microsoft may also develop its own Arm server chips.
New U.S. export restrictions targeting high‑end AI GPUs such as Nvidia’s H800 and A800 aim to curb China’s access to advanced compute, potentially slowing its AI model development, affecting major chip makers and prompting Chinese firms to stockpile hardware or accelerate domestic chip efforts.
The article reviews Baidu's 2023 generative AI initiatives, details the inaugural Baidu Commercial AI Technology Innovation Competition co‑hosted with the China AI Society and NVIDIA, highlights winning teams' technical approaches in conversion prediction and inference optimization, and shares insights from industry leaders on future AI talent and innovation.
This article provides a comprehensive introduction to NVIDIA's NeMo toolkit for conversational AI, detailing its ASR capabilities, model architectures, training workflow, Chinese language support, deployment options, and additional speech AI features such as VAD and speaker diarization.
The article outlines NVIDIA’s CLARA platform, BioNeMo framework, and GPU‑accelerated tools such as CLARA Parabricks and RAPIDS, demonstrating how AI and high‑performance computing dramatically speed up drug‑target identification, molecular generation, protein structure prediction, and high‑throughput DNA/RNA sequencing, with benchmarks showing up to 80‑fold acceleration.
This guide explains how to set up NVIDIA drivers, install the NVIDIA device plugin, and create a Kubernetes pod that requests GPU resources, providing step‑by‑step commands and a sample YAML manifest for GPU‑enabled workloads.
