EDN

Round pegs, square holes: Why GPGPUs are an architectural mismatch for modern LLMs

The saying “round pegs do not fit square holes” persists because it captures a deep engineering reality: inefficiency most ...

Is this the 'Nvidia killer'? Cerebras’ wafer-scale AI engine sparks supremacy showdown

Dubbed the "Nvidia killer," Cerebras' wafer-scale engine has reportedly crushed Nvidia's H200 in raw AI training power: 125 ...
SDxCentral

Beyond CUDA: Inside the push to loosen Nvidia’s grip on AI computing

As the AI arms race intensifies and the costs of vendor lock-in rise, a new class of challengers is stepping into the ring to loosen Nvidia’s grip on AI computing. Legacy tech companies such as AMD ...
kr-asia

Iluvatar CoreX becomes first Chinese GPGPU firm to gain HKEX listing approval

Image source: Iluvatar CoreX. The Chinese chipmaker has built a vertically integrated GPGPU ecosystem, and now seeks new capital to scale it. Shanghai-based semiconductor company Iluvatar CoreX has ...
Forbes

Graphics Processing Units: From Pixels To Parameters

Shashank is the Chief AI Officer at MooresLab AI, where he helps chip companies leverage Agentic AI to accelerate silicon engineering tasks. From training large language models with hundreds of ...
Digi Times

AI hardware landscape shifts from GPUs to NPUs as edge computing gains ground

As artificial intelligence (AI) transitions from cloud-centric operations to edge devices, the hardware that powers AI is undergoing significant change. While graphics processing units (GPUs) have ...
New Atlas

Single-shot light-speed computing might replace GPUs

Want to call someone a quick-thinker? The easiest cliché for doing so is calling her a computer – in fact, “computers” was the literal job title of the “Hidden Figures” mathematicians who drove the ...
kr-asia

Moore Threads makes record IPO debut as China’s AI chip fever rages on

The MTT S4000, a GPU designed by Moore Threads. Image source: Moore Threads. The firm’s valuation appears stretched, with investor enthusiasm outpacing its underlying financials. As demand for ...
The Conversation

Google is relying on its own chips for its AI system Gemini. Here’s why that’s a seismic change for the industry

Alaa Mohasseb does not work for, consult, own shares in or receive funding from any company or organization that would benefit from this article, and has disclosed no relevant affiliations beyond ...
CNBC

Nvidia takes $2 billion stake in Synopsys with expanded computing power partnership

Nvidia on Monday announced it has purchased $2 billion of Synopsys' common stock at $414.79 per share. The two companies are partnering to accelerate computing and AI engineering solutions. "This is a ...
Digi Times

Nvidia defends GPU lead as Google TPU gains traction in AI chip market

Nvidia has asserted that its graphics processing unit (GPU) platform remains a full generation ahead of its competitors, responding to increased attention on Google's Tensor Processing Unit (TPU) in ...