The Next Platform

Nvidia has spiced up its 2028 AI hardware roadmap with reveals about the future Rosa-Feynman systems. $NVDA nextplatform.com/compute/2026/0…

@Matt_Pinner Humanity. <Smile>

In your opinion, what's the biggest threat to humanity?

Agentic AI will require all kinds of compute engines, including CPUs, GPUs, DPUs, and now LPUs. $NVDA $GOOG $AMD $INTC nextplatform.com/ai/2026/03/17/…

Is AI networking upstart Eridu working on a high radix, waferscale switch ASIC? Maybe. Just maybe. $NVDA $AMD $AVGO $MRVL nextplatform.com/connect/2026/0…

How Ayar Labs is going to ramp silicon photonics into the 2028 generation of AI systems. $NVDA $AMD nextplatform.com/connect/2026/0…

Is Nvidia looking to add optical circuit switches to its AI infrastructure? $NVDA $LITE $COHR nextplatform.com/connect/2026/0…

SambaNova says its SN50 systems can best the tokenomics of Nvidia GPUs for agentic AI $NVDA $AMD $GOOG $AWS nextplatform.com/ai/2026/02/25/…

It looks like Meta is going to spend almost as much money on AMD GPUs as it does for Nvidia GPUs. $META $AMD $NVDA nextplatform.com/2026/02/24/som…

Meta Platforms is getting access to millions of Nvidia GPUs – don’t assume it is buying all of them. $NVDA $META nextplatform.com/2026/02/18/som…

@curiosityonx Me.

If aliens came to Earth, who should be in charge of speaking on behalf of humanity?

Taalas etches AI model weights in ROM with big SRAM memory, claims 50X more performance at 1/50th the cost per token with 1/36th the latency of a GPU – at 1/6th the energy. nextplatform.com/2026/02/19/taa…

IBM banks on lots of on-premises, small LLM models being used for GenAI inference by enterprises. $IBM $MSFT $NVDA $AMD $GOOG $AMZN nextplatform.com/2026/01/29/big…

The new “Braga” Maia 200 AI XPUs from Microsoft – big improvement, bigger clusters. Will OpenAI use it? Will Azure rent it out? $MSFT $GOOG $NVDA $AMD nextplatform.com/2026/01/28/mic…

Intel has to choose between meeting demand for high-end servers or high-end PCs. $INTC $AMD $NVDA $TSM nextplatform.com/2026/01/23/int…

Who will be the first to ship a high radix, low latency UALink switch? It just might be Upscale AI. $NVDA $AMD $AVGO $CSCO nextplatform.com/2026/01/21/ups…

TSMC says the GenAI Boom goes out for years, but knows its forecasts are only as good as those of its customers. $TSM $NVDA $AMD $GOOG $MSFT $AMZN nextplatform.com/2026/01/16/tsm…

Cerebras got a $10 billion cloud deal with OpenAI, and Groq didn’t and got acquihired by Nvidia. Correlate that. $NVDA $AMD $GOOG nextplatform.com/2026/01/15/cer…

Sorry this took so long, but we finally got our deep dive on the Rebellions AI Rebel inference chips done. Happy holidays everyone, see you in 2026! nextplatform.com/2025/12/23/reb…

@sciencegirl A freaking nightmare?

A group of rats is called?

Micron’s HBM business is up by 5.7X in the latest quarter, and is going to be the company’s dominant product before too long. $MU $NVDA nextplatform.com/2025/12/19/hbm…