It cuts energy consumption by 70% compared to 2nm.
OMG @IBM just announced its 0.7nm process node using homegrown NanoStack technology. It defines a route for transistors after GAA by using a staggered CFET design enabling 50% logic scaling, 40% SRAM scaling, and 666 million transistors per sq mm***.
https://morethanmoore.substack.com/p/ibms-announces-07nm-process-node
Many users expressed excitement over IBM's 0.7nm NanoStack CFET chip and 3D transistor stacking because of major SRAM scaling gains, high transistor density, and signs of renewed competitiveness against TSMC and Intel.
No Digg Deeper questions have been answered for this story yet.
IBM just unveiled a sub-1 nanometer chip breakthrough.
That honestly wasnt on my bingo card.
Its new 0.7 nm / 7 angstrom technology uses a 3D "nanostack" transistor architecture to vertically stack and stagger transistors.
IBM says it can fit nearly 100 billion transistors onto a chip the size of a fingernail, almost 2x the density of its 2 nm chip from 2021.
• up to 50% more performance • or 70% better energy efficiency • 40% SRAM scaling for AI workloads
Important caveat: this is still research, not a chip shipping tomorrow. IBM says production could happen as early as the next 5 years.
The world’s first sub‑1 nanometer node chip is here.
Delivering 70% greater energy efficiency, this breakthrough powers a new era of computing that’s more capable while using less energy.
Dig into this next-gen tech: https://ibm.co/6016EOHpM
IBM debuts world’s first sub-1 nanometer chip.
They just announced a 0.7nm chip technology.
(0.7nm is about 100,000 times thinner than a human hair, only a few atoms across as a technology marker,)
It stops treating smaller chips as a flat-layout problem and starts stacking transistors upward through a new nanostack architecture.
A chip node no longer means every feature has that exact size, but 7 angstroms means IBM is pushing transistor design into a scale where individual atoms start to shape the engineering limits.
Traditional scaling packs more switches side by side, while nanostack vertically staggers nanosheet transistors, letting each layer use different materials for speed or lower power instead of forcing one transistor recipe across the chip.
IBM says the design can fit nearly 100B transistors on a fingernail-sized chip, deliver up to 50% more performance or 70% better energy efficiency than its 2nm node, and shrink SRAM by 40%, which is huge because AI chips constantly wait on memory.
Practically it means future chips that can run bigger AI models, phones, laptops, servers, and cloud systems faster while using much less power, because more compute and memory can fit into the same tiny chip area.
---
newsroom. ibm. com/2026-06-25-ibm-debuts-worlds-first-sub-1-nanometer-chip-technology
@LordStreetGuru @IBM They're still patenting processes and licensing them.
> at the chip store > ask about their current process node > "oh it's sub 1-nm" > surprised, ask whether that means the transistor pitch is < 1nm > they don't understand > show them a diagram, explain what I mean > "It's a good chip sir." > buy chip, take it to lab. > put it under microscope, look at transistors > 90 nm wide. 16nm pitch. >MFW 😑
IBM just unveiled a sub-1 nanometer chip breakthrough.
That honestly wasnt on my bingo card.
Its new 0.7 nm / 7 angstrom technology uses a 3D "nanostack" transistor architecture to vertically stack and stagger transistors.
IBM says it can fit nearly 100 billion transistors onto a chip the size of a fingernail, almost 2x the density of its 2 nm chip from 2021.
• up to 50% more performance • or 70% better energy efficiency • 40% SRAM scaling for AI workloads
Important caveat: this is still research, not a chip shipping tomorrow. IBM says production could happen as early as the next 5 years.
@TableFox @IBM No. ASML make equipment, IBM develop transistors and processes.
@highyieldYT @IBM IBM Research licensed 2nm to Rapidus. IBM Systems (Power/Z) use Samsung. Different BUs.
@IanCutress @IBM Ian, does this mean IBM is a future competitor of ASML?
@Mads_Skipper @IBM Well consider they announced GAA back in 2021 and we only saw the first chips come out in ultra late 2025. We've still got plenty of runway with generations of GAA to go, at least 3-4 more.
@IanCutress @IBM Does IBM still work with GloFo or who is their foundry partner?
@IanCutress @IBM IBM is still making things?
@IanCutress @IBM Their marketing should call it the process node of the beast
@Eveosian @LordStreetGuru @IBM In this case, just licensing. QPUs need a lot of classical compute, but won't use CFET themselves.
@i2cjak For a proof of concept? More than enough.
IBM just debuted what it says is the world’s first sub 1 nanometer chip technology at the 0.7 nm node.
I genuinely thought getting below 1 nm was still a 2028 type problem at best, especially with all the weird materials, scaling limits, and post silicon architecture questions people expected around carbon nanotubes.
“Built with revolutionary nanostack 3D chip architecture, IBM’s sub 1 nm chip to propel semiconductor industry forward for the next decade.”
IBM says the chip can pack nearly 100 billion transistors onto something the size of a fingernail, with up to 50 percent more performance or 70 percent better energy efficiency than its 2 nm node.
The mass manufacturing is still a few years out.
IBM says it sees a path to production in as early as the next 5 years, so earliest adoption would likely be around 2031 if the roadmap holds.
@i2cjak Actual experts seem to be more optimistic
@Never_Hedgee nothing about this seems especially optimistic, the article you linked seems to say that yield is still a big open problem
@RrichPRMR yes, 100%. But at the same time, it is also necessary. After all, chips must not only become better, but also more efficient so that development can continue.
@IanCutress @LordStreetGuru @IBM Will they be sticking to just licensing to companies like Rapidus or will they be manufacturing themselves with Anderon to facilitate developing traditional GPUs, CPUs along with their signature QPUs
@kimmonismus this makes the next few years look really interesting ))
@i2cjak @mattparlmer very optimistic I see!
