More:
A whole new line. Three Windows Machines.
Max Weinbach@mweinbach
Here are the RTX Spark specs
Full Blackwell GPU, up to 128GB of memory, 1 Petaflop of FP4 compute
9:33 PM · May 31, 2026 · 1.5K Views
NVIDIA and Microsoft announce RTX Spark, a consumer ARM superchip combining a Blackwell GPU and Grace CPU
AI Judge changed title after evaluation, original title: "NVIDIA and Microsoft launch RTX Spark PCs and N1X hardware to run local AI agents at 1 petaflop"
The platform delivers one petaflop of FP4 AI performance
1K12.4K9321.8K3.9M
Sentiment
Many users expressed excitement about NVIDIA's RTX Spark superchip bringing petaflop AI performance to personal laptops and PCs, while some dismissed it as overpriced or criticized Windows integration.
Pos
67.4%
Neg
32.6%
215 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Posts from X
Most Activity
Most Activity
VIEWS208.2KBOOKMARKS304LIKES1.7KRETWEETS96REPLIES85
Chubby♨️@kimmonismus
Holy, NVIDIA revealed RTX Spark and its nuts. Superchip. Blackwell-GPU (RTX-5070-Cores) 128GB, 1 PetaFLOP KI, 1440p/100FPSand so much for Laptops. Insane
28dViews 208.2KLikes 1.7KBookmarks 304
Chubby♨️@kimmonismus
At Computex, NVIDIA walked into a market it never owned: the PC itself.
RTX Spark is an ARM superchip: 20-core Grace CPU, Blackwell GPU with 6,144 CUDA cores, up to 128GB unified memory, 1 PetaFLOP FP4. The GPU core count sits at RTX 5070 (!) level, in a laptop as thin as a MacBook.
However, the strategy is the story:
NVIDIA used to sell graphics cards. Now it builds the whole system: CPU, GPU, memory and software stack on one package, all on Windows on ARM. That puts it head to head with Intel, AMD, Qualcomm and Apple at the same time. Four rivals, one chip.
The framing is as important as the gaming demos, even with Jensen holding up two laptops running 007 First Light and Forza Horizon 6. NVIDIA calls this a "personal AI computer." The pitch: your PC becomes a teammate, with local agents working next to you while you do something else. 128GB unified memory means 120B-parameter models running locally, no cloud.
NVIDIA is wagering that the PC of the next few years gets built around local AI (!) rather than apps. If that lands, the company has done more than ship a product. It's set the frame everyone else has to compete in. What's still open: the GPU is power-limited (45–80W), the FPS and battery claims are unbenchmarked, and Windows on ARM still drags a long list of gaming and x86-emulation problems behind it. Ships Fall 2026. Until then, these are slides, not reviews. This is the most interesting AI launch of the year is a chip.
28dViews 27.8KLikes 492Bookmarks 83
Ryan Shrout@ryanshrout
Spent some time with @Nvidia and @Microsoft on the new @Surface Laptop Ultra with RTX Spark.
This is a hell of a machine for sure; going to be interesting to see how its performance scales and how the software story plays out. Early demos were impressive.
28dViews 45.1KLikes 567Bookmarks 82
@jason@Jason
Image dump for may 1st 2026
28dViews 98.4KLikes 252Bookmarks 42
whoah - Grace + Blackwell chips in a laptop.
@Microsoft + @NVIDIA teaming up to take on 6 years of total dominance of Apple Silicon
28dViews 37KLikes 154Bookmarks 33
Chubby♨️@kimmonismus
This is NVIDIAs/ Microsofts attack on Mid Tier MacBook Pros with M-Pro Chips.
Makes me even more excited for Apples M6. Competition is heating up friends!
28dViews 19.8KLikes 238Bookmarks 25
MTS@MTSlive
SITUATION EXPLAINED: On device compute means infinitely free tokens.
We asked @stevesi, who built the Surface and led Windows at Microsoft, why local compute is the inevitable next step for AI:
"The problem is that everybody is gated by the consumption of tokens, which cost money, or you can't get them if you're trying to use them for free."
"If you just wanna let something roll for three days you really don't wanna end up with a $10,000 bill, that's why people are stacking Mac Minis."
"Any time there's a resource constraint that you have to pay for, it moves to your device and becomes free. I just don't imagine how it can't happen."
MTS@MTSlive
SITUATION DETECTED: Nvidia has announced RTX Spark, its first consumer chip for Windows laptops and desktops.
The ARM-based superchip combines a 20-core Grace CPU, Blackwell GPU, and 128GB unified memory in a single package. Coming Fall 2026.
28dViews 27.2KLikes 101Bookmarks 38
NVIDIA@nvidia
Create, build, and play: https://nvidianews.nvidia.com/news/nvidia-microsoft-windows-pcs-agents-rtx-spark
29dViews 31.1KLikes 137Bookmarks 24
WKeyGod@WKeyGod420
@nvidia No one gives a fuck. Give us 1080ti performance for 700 dollars.
29dViews 6KLikes 91
AlphaFox@alphafox
@nvidia Who is this even marketed at? Nobody I know.
29dViews 5.7KLikes 43
Steven Sinofsky@stevesi
Hey there...gonna be on @MTSlive at 12:30PT to talk about this. Very exciting. Join in!
MTS@MTSlive
SITUATION DETECTED: Nvidia has announced RTX Spark, its first consumer chip for Windows laptops and desktops.
The ARM-based superchip combines a 20-core Grace CPU, Blackwell GPU, and 128GB unified memory in a single package. Coming Fall 2026.
28dViews 10.4KLikes 30Bookmarks 0
@jason@Jason
@jason@Jason
28dViews 18.6KLikes 16Bookmarks 3
HealthRanger@HealthRanger
@ns123abc Slower than molasses. The tokens / second output will be in single digits for most of the dense models people actually want to run. Effective 273GB/s unified memory bandwidth is a joke.
28dViews 719Likes 15Bookmarks 1
Microsoft Surface@surface
@ryanshrout It belongs in the hands of world makers like you.
28dViews 1.2KLikes 17
kumikumi (Ankkala)@ankkala
@nvidia every time someone tries to make a macbook competitor, they make it run windows
28dViews 574Likes 16
Irfan Ahmad@Irfuu_
@nvidia Jensen Huang says the AI PC reinvention is as big as the smartphone shift by calling it “a new line” and “a new beginning.”
29dViews 456Likes 6Bookmarks 2
GentleScotch@scotchdrunked
@WKeyGod420 @nvidia Are you stupid? RTX 5060 costs 300-400 USD and beats the 1080ti.
29dViews 664Likes 20
ScreamerRSA@ScreamerRSA
@nvidia One step closer to "you will one nothing and you will pay us a subscription to generate AI models for us".
29dViews 655Likes 8Bookmarks 1
Alps@alpaysh
@ns123abc long $NVDA
29dViews 726Likes 9Bookmarks 1
Sebastian C@SebTH1
@nvidia We dont want slop chips regular people don't want AI Gamers especially don't.
29dViews 1.9KLikes 7
Load more posts