#nvlink

@llms.activitypub.awakari.com.ap.brid.gy

5 days ago

Forget About Chips. It’s the System That Matters For AI Picking the right processor for a particular workload will always be important. But when it comes to AI–and more specifically, AI inferen...

#Features #chiplet #GPUs #HBM #interconnect #NVLink #Fusion #PCI […]

[Original post on hpcwire.com]

0 0 0 0

LLMs

@llms.activitypub.awakari.com.ap.brid.gy

5 days ago

Original post on hpcwire.com

Forget About Chips. It’s the System That Matters For AI Picking the right processor for a particular workload will always be important. But when it comes to AI–and more specifically, AI inferen...

#Features #chiplet #GPUs #HBM #interconnect #NVLink #Fusion #PCI […]

[Original post on hpcwire.com]

0 0 0 0

Ryan Daws

@gadgetry.bsky.social

1 week ago

NVIDIA and Marvell alliance scales AI-RAN infrastructure NVIDIA and Marvell Technology have formed a partnership designed to link Marvell into the NVIDIA AI factory and the wider AI-RAN ecosystem.

Advancements in optical interconnects directly address the growing need for high-speed, low-latency networking architectures necessary to support heavy AI workloads at the edge. #nvidia #marvell #airan #telecoms #nvlink #5g #6g #ai #technology

0 0 0 0

@soniccomponents.bsky.social

2 weeks ago

www.soniccomponents.com/nvidia-dgx-r...
Pre order yours today. NVIDIA DGX Rubin NVL8 #NVIDIA #DGXRubin #NVL8 #AIInfrastructure #Supercomputing #GPUs #NVLink #DeepLearning #MachineLearning #LLMs #AIHardware #DataCenters #HighPerformanceComputing #HPC #AIRevolution #NextGenAI #TechInnovation

1 0 0 0

Meng Li

@mengli512.bsky.social

4 weeks ago

SerDes, More Critical Than Ever SerDes is the key technology enabling high-speed data transmission in AI computing. From Broadcom and Marvell to MediaTek, mastering SerDes determines scalability and system performance in the AI era.

SerDes is now the backbone of AI scalability. From Broadcom to MediaTek, ASIC dominance depends on it. NVLink, UALink, and startups like Credo are all racing on 224G/448G.
#SerDes #AI #NVLink #UALink #CPO #optical
open.substack.com/pub/chippub/...

0 0 0 0

ServeTheHome Server Storage and Networking Reviews

@servethehome.com.web.brid.gy

2 months ago

SiFive To Adopt NVLInk Fusion For Future Data Center RISC-V CPU Designs For a while now, NVIDIA has been working to turn its NVLink high speed interconnect technology into more than just a bus used by NVIDIA, aiming to fashion an entire ecosystem around it. Those efforts kicked into high gear last year with the announcement of NVLInk Fusion, a program to allow third parties to connect […] The post SiFive To Adopt NVLInk Fusion For Future Data Center RISC-V CPU Designs appeared first on ServeTheHome.

1 0 1 0

Cool3c

@cool3c.com.web.brid.gy

3 months ago

NVIDIA Rubin超級運算平台以6大晶片、軟硬體協同設計降低10倍推論詞元成本、僅需1/4的GPU即可訓練混合專家模型 隨著NVIDIA預期在2026年3月的GTC大會正式揭開Rubin平台細節，NVIDIA執行長黃仁勳於CES 2026的主題演講尾聲繼續聚焦Rubin平台，強調透過全新6款晶片高度的協同設計，能大幅縮減訓練時間與降低推論詞元成本，相對Blackwell可將推論詞元(Token)成本降低10倍、並僅需1/4的GPU即可完成混合專家模型訓練。6大全新晶片與2種平台▲Rubin平台由6款緊密協同設計的新晶片構成平台基礎黃仁勳強調，Rubin的平台進一步將關鍵元件進行極致的協同設計，使每個晶片在其負責的領域發揮極致的性能，進一步突破性能極限，相較Blackwell又是一次更重大的技術突破。Rubin平

0 0 0 0

handle.invalid

@handle.invalid

3 months ago

NVIDIA「Rubin」が量産突入へ：Blackwellを過去にする「50ペタフロップス」の衝撃とJensen Huangの加速戦略 2026年のCES（Consumer Electronics Show）において、AIコンピューティングの覇者NVIDIAは、次世代アーキテクチャ「Vera Rubin」プラットフォームを正式に発表した。業界の予想を大 ... Read more

0 0 0 0

Cool3c

@cool3c.com.web.brid.gy

4 months ago

NVIDIA 2026年GTC訂於3月16日至19日，同樣選在加州聖荷西舉辦 NVIDIA GTC 2026 訂於 3/16-19 舉行，將發表次代 Vera CPU 與 Rubin GPU 架構，聚焦物理 AI 等創新。大會提供實體及免費線上方案，供全球開發者參與。

0 0 0 0

NEWSTECNICAS | Tecnología , IA y Gaming

@newstecnicas.info.ve

4 months ago

AWS Integra el NVLink Fusion de Nvidia en su Chip Trainium4 para Construir 'Fábricas de IA' Ultrarrápidas y Desafiar a Google

#AWS Integra el #NVLink Fusion de #Nvidia en su Chip #Trainium4 para Construir 'Fábricas de IA' Ultrarrápidas y Desafiar a Google www.newstecnicas.info.ve/2025/12/aws-...

0 0 0 0

handle.invalid

@handle.invalid

4 months ago

NVIDIA、Blackwellが前世代比10倍の性能を記録と発表：GB200 NVL72が「Mixture of Experts」で叩き出した性能の衝撃とは 2025年12月4日、NVIDIAは同社の最新鋭AIプラットフォーム「Blackwell」アーキテクチャ、具体的にはラック・スケール・システムであるGB200 NVL72が、現在のAI開発の主流であるMixture of ... Read more

0 0 0 0

Gerrit Eicker

@eicker.bsky.social

4 months ago

#AWS announced #Trainium3, a new #AItrainingchip with significant performance and energy efficiency improvements. #Trainium4, already in development, will offer even better performance and support #Nvidia’s #NVLink Fusion technology, potentially attracting more AI applications to AWS.…

0 0 0 0

DeadStack / Technology

@deadstack.net

4 months ago

Tues, December 02, 2025: OpenAI goes CODE RED. Amazon puts NVLink in new chips & launches an AI agent suite at re:Invent. Mistral OSS models. Tuesday, December 02, 2025 Summary OpenAI declares code red on competitive model quality. Amazon is reshaping its AI strategy with dual announcements.

🤯 Amazon comes out SWINGING in AI.

Tues, December 02, 2025: OpenAI declares CODE RED, #Amazon puts #NVLink in new chips & launches an AI agent suite at re:Invent. #Mistral OSS models, & #ChatGPT + Apple Health

www.linkedin.com/pulse/tues-d...

1 0 0 0

ServeTheHome Server Storage and Networking Reviews

@servethehome.com.web.brid.gy

4 months ago

NVIDIA NVLink Fusion Tapped for Future AWS Trainium4 Deployments NVIDIA NVLink Fusion has been tapped for future AWS Trainium4 deployments as AWS selects NVIDIA for scale-up capabilities The post NVIDIA NVLink Fusion Tapped for Future AWS Trainium4 Deployments appeared first on ServeTheHome.

0 0 0 0

NVIDIA Blog

@blogs.nvidia.com.web.brid.gy

4 months ago

NVIDIA Partners With Mistral AI to Accelerate New Family of Open Models Today, Mistral AI announced the Mistral 3 family of open-source multilingual, multimodal models, optimized across NVIDIA supercomputing and edge platforms. Mistral Large 3 is a mixture-of-experts (MoE) model — instead of firing up every neuron for every token, it only activates the parts of the model with the most impact. The result is efficiency Read Article

0 0 0 0

Robert Mooney

@vermelho18.bsky.social

4 months ago

#Amazon #Trainium #AWS #Nvidia #NVLink

0 0 0 0

Home – Blocks and Files

@blocksandfiles.com.web.brid.gy

4 months ago

CXL 4.0 doubles bandwidth and stretches memory pooling to multi-rack setups The fourth CXL memory sharing spec adds speed and extends distance for multi-rack systems. The Computer eXpress Link (CXL) specification defines how pools of memory can be connected across the PCIe bus. There have been three generations. CXL 1 enabled x86 servers to access PCIe 5.0-linked memory in external devices like SmartNICs and DPUs. CXL […]

0 0 0 0

Cool3c

@cool3c.com.web.brid.gy

4 months ago

Arm Neoverse 平台設計將整合 NVIDIA NVLink Fusion 加速異構運算佈局 Arm Neoverse 平台將整合 NVIDIA NVLink Fusion，透過 AMBA CHI C2C 介面，實現 CPU 與 GPU 間高頻寬一致性互連。此舉旨在加速 AI 異構運算，為採用 Neoverse 設計自研 CPU 的客戶，提供更高效能與架構彈性。

0 0 0 0

Ahmandonk

@ahmandonk.bsky.social

4 months ago

📰 Arm Perluas Platform Neoverse dengan Integrasi NVIDIA NVLink Fusion

👉 Baca artikel lengkap di sini: ahmandonk.com/2025/11/19/arm-neoverse-...

#ai-infrastructure #arm #datacenter #neoverse #nvidia #nvlink

0 0 0 0

Gerrit Eicker

@eicker.bsky.social

4 months ago

#Arm and #Nvidia are #partnering to integrate Arm-based #Neoverse #CPUs with Nvidia’s #GPUs using Nvidia’s #NVLink Fusion technology. This #collaboration will benefit customers, particularly #hyperscalers, who prefer custom infrastructure setups. The partnership highlights Nvidia’s strategy of…

0 0 0 0

ServeTheHome Server Storage and Networking Reviews

@servethehome.com.web.brid.gy

4 months ago

Arm Joins the NVIDIA NVLink Fusion Ecosystem Arm joins the NVIDIA NVLink Fusion ecosystem, giving custom Arm CPU designers an easier pathway to integrate NVIDIA GPUs The post Arm Joins the NVIDIA NVLink Fusion Ecosystem appeared first on ServeTheHome.

0 0 0 0

handle.invalid

@handle.invalid

4 months ago

Arm Neoverseプラットフォームが NVIDIA NVLink Fusionを統合：NVIDIAの支配は更に強固な物となるのか？ AIが社会のあらゆる構造を変革する現代において、その心臓部であるデータセンターのアーキテクチャもまた、世代に一度の大きな転換点を迎えている。この変革の主戦場となっているのが、膨大なデータを処理するCPUとGPUをいかに効 ... Read more

0 0 0 0

@soniccomponents.bsky.social

5 months ago

www.soniccomponents.com/nvidia-black...
Unlocking Real-Time Trillion-Parameter Models
GB200 NVL72 connects 36 Grace CPUs and 72 Blackwell GPUs in liquid-cooled design. It boasts a 72-GPU NVIDIA NVLink™ , massive GPU@nvidia.bsky.social #gb200 #nvl72 #36grace #cpu #72blackwell #gpu #nvlink

0 0 0 0

Ahmandonk

@ahmandonk.bsky.social

5 months ago

📰 NVIDIA Pamerkan Wujud Perdana “Vera Rubin” Superchip dengan 100 PetaFLOPS FP4 Compute

👉 Baca artikel lengkap di sini: ahmandonk.com/2025/10/30/nvidia-vera-r...

#ai #training #cpu #arm #exascale #gpu #hbm4 #hpc #nv144 #nvidia #nvlink #nvlink #1.8tbps #supe

0 1 0 0

Ahmandonk

@ahmandonk.bsky.social

5 months ago

📰 NVIDIA “Vera Rubin” NVL144 Siap Produksi Massal pada 2026, Gandakan Kepadatan Komputasi AI

👉 Baca artikel lengkap di sini: ahmandonk.com/2025/10/21/nvidia-vera-r...

#ai #server #blackwell #foxconn #gpu #hbm4 #nvidia #nvl144 #nvlink #6 #vera #rubin

0 0 0 0

AndreevWebStudio

@andreevwebstudio.com

5 months ago

Meet the next wave of AI infrastructure: NVIDIA GB200 NVL72. Think rack-scale “super-GPU” built for AI factories—30× faster inference,
Read the full post: aiinovationhub.com/nvidia-gb200...

#NVIDIA #GB200 #NVL72 #AIFactories #LLM #Inference #NVLink #HBM #Blackwell #aiinnovationhub