alt.hn

3/16/2026 at 8:01:02 PM

Nvidia Launches Vera CPU, Purpose-Built for Agentic AI

 https://nvidianews.nvidia.com/news/nvidia-launches-vera-cpu-purpose-built-for-agentic-ai

by lewismenelaws

3/17/2026 at 12:05:08 AM

Agentic AI CPU? No.

It’s a CPU designed for an AI cluster. Their last CPU Grace was the same thing and no one called it agentic.

Vera now just has more performance/more bandwidth. It’s cool, I’d like to have one of these clusters, but this is not new.

It’s marketed as agentic AI because that’s fashionable in 2026.

by WhitneyLand

3/17/2026 at 12:21:58 AM

They significantly lowered latency compared to EPYC/Xeon, which is critical for streaming agents (e.g. text/audio/video agents).

by storus

3/17/2026 at 2:50:44 AM

What latency? How much is it compared to LLM inference speed?

by stingraycharles

3/17/2026 at 1:46:47 PM

See the Redpanda comment/link here.

by storus

3/16/2026 at 8:09:42 PM

Given the price of these systems the ridiculously expensive network cards isn't such a huge huge deal, but I can't help but wonder at the absurdly amazing bandwidth hanging off Vera, the amazing brags about "7x more bandwidth than pcie gen 6" (amazing), but then having to go to pcie to network to chat with anyone else. It might be 800Gbe but it's still so many hops, pcie is weighty.

I keep expecting we see fabric gains, see something where the host chip has a better way to talk to other host chips.

It's hard to deny the advantages of central switching as something easy & effective to build, but reciprocally the amazing high radix systems Google has been building have just been amazing. Microsoft Mia 200 did a gobsmacking amount of Ethernet on chip 2.8Tbps, but it's still feels so little, like such a bare start. For reference pcie6 x16 is a bit shy of 1Tbps, vaguely ~45 ish lanes of that.

It will be interesting to see what other bandwidth massive workloads evolve over time. Or if this throughout era all really ends up serving AI alone. Hoping CXL or someone else slims down the overhead and latency of attachment, soon-ish.

Maia 200: https://www.techpowerup.com/345639/microsoft-introduces-its-...

by jauntywundrkind

3/16/2026 at 8:31:21 PM

> It might be 800Gbe but it's still so many hops, pcie is weighty.

Once you need to reach beyond L2/L3 it is often the case that perfectly viable experiments cannot be executed in reasonable timeframes anymore. The current machine learning paradigm isn't that latency sensitive, but there are other paradigms that can't be parallelized in the same way and are very sensitive to latency.

by bob1029

3/16/2026 at 8:28:31 PM

Most of the big AI/HPC clusters these systems are aimed at aren’t running regular PCIe Ethernet between nodes, they’re usually wired up with InfiniBand fabrics (HDR/NDR now, XDR soon)

by babelfish

3/17/2026 at 4:31:23 AM

Infiniband cards are connected to the rest of the machine with the PCIe as well

by tryauuum

3/20/2026 at 11:57:40 AM

I think you're right - the Tauri vs Electron comparison isn't quite the same scale of difference.

Both still run web tech in a wrapper, just with different performance characteristics. The local-first vs cloud distinction is more fundamental, especially for tools that interact with platforms like LinkedIn.

When I built ZenMode, the core insight was that LinkedIn can easily detect automation coming from AWS/datacenter IPs, but when your desktop app uses your actual Chrome browser and home IP, it's indistinguishable from manual usage.

That's why we went with an Electron/Puppeteer architecture running locally rather than yet another cloud service. Check it out at https//zen-mode.io if you're curious about the local execution model.

by jal505

3/16/2026 at 9:18:05 PM

Say what you want about NVIDIA (to me they are just doing what every company would do in their place), but they create engineering marvels.

by baal80spam

3/16/2026 at 11:59:56 PM

[flagged]

by szmarczak

3/17/2026 at 12:11:26 AM

Who is NVIDIA killing, exactly?

Is Apple complicit in killings because operators planned missions on Macbooks? Dell? Microsoft?

by jazzpush2

3/17/2026 at 7:05:13 AM

Whom, not who.

by penguin_booze

3/17/2026 at 7:52:03 AM

[flagged]

by szmarczak

3/17/2026 at 8:12:23 AM

> You're lacking all the brain cells.

You can't comment like this on Hacker News, no matter what you're replying to.

Please take a moment to read the guidelines and make an effort to observe them in future.

https://news.ycombinator.com/newsguidelines.html

by tomhow

3/17/2026 at 8:27:06 AM

Immoral behavior needs to be called out. See https://news.ycombinator.com/item?id=47274075

by szmarczak

3/17/2026 at 9:38:07 AM

The guidelines don’t get disregarded just because the topic is important to you. People have been trying that trick since the beginning. The guidelines are only worth having if they enable us to discuss difficult topics without burning this place to the ground. If you’re not able to discuss a topic without making swipes about someone’s “brain cells”, you need to find a different discussion forum. This site is only here for you to post that kind of trash because other people make the effort to raise the standards rather than drag them down.

by tomhow

3/17/2026 at 12:12:19 AM

GE, Samsung, Microsoft, Google, IBM, and so many others

by zoklet-enjoyer

3/16/2026 at 8:17:16 PM

It is a 88-core ARM v9 chip, for somewhat more detailed spec.

by d_silin

3/16/2026 at 9:54:01 PM

Vera does what NVIDIA calls Spatial Multithreading, "physically partitioning each core’s resources rather than time slicing them, allowing the system to optimize for performance or density at runtime." A kind of static hyperthreading; you get two threads per core.

It's somewhat different from how x86 chips do simultaneous multithreading (SMT),

by PeterCorless

3/17/2026 at 5:35:28 PM

Seems like curious terminology from NV. In estabilished use, SMT means executing instructions from several cpu threads concurrently in the OOO CPU's execution units so they are not starved from work, whereas timeslicing conventionally means context switching between threads/processes, alternating temporally.

In operating systems timeslicing means giving a quantum of execution time to each process, and context switching between processes. Not normally a term used in computer architecture but possibly the characterisation would fit a barrer processor rather than SMT.

by fulafel

3/16/2026 at 9:04:53 PM

Hmm, the 128-core Ampere Altra CPU is already available, and in a case from System76. I wonder what else differentiates it.

If they're going to build CPUs I wish they had used Risc-V instead. They are using it somewhat already.

by mixmastamyk

3/16/2026 at 10:06:10 PM

You can see here[1] what the specs are for the CPU (listed as "NVIDIA Vera Rubin Superchip").

The CPU is integrated with two Rubin GPUs but each of the CPU cores has dedicated FP8 acceleration as well.

1. https://www.nvidia.com/en-us/data-center/vera-rubin-nvl72/

by OneDeuxTriSeiGo

3/17/2026 at 4:31:04 AM

I own one of these systems. My interpretation is the Ampere systems are targeted at lower cost scale out. The Ampere Altra CPUs are limited to DDR4. The raw single core performance doesn’t match Intel or AMD offerings. You get a lot of cores for a lower hardware cost and at lower energy usage.

The Nvidia CPUs are designed for a very specific use case. They are designed for high performance with less concern about cost control.

The newer AmpereOne CPUs use DDR5 with the AmpereOne M supporting even higher memory bandwidth. Even then, I doubt the AmpereOne CPUs will match the performance of the Nvidia Rubin CPUs. But the Ampere processors are available for general use. I am guessing that Nvidia is only going to sell the complete rack system and only to high-volume customers.

by ibgeek

3/16/2026 at 9:08:08 PM

Anyone know how this compares to Apple’s M5 chips? Or is that comparison <takes off sunglasses> apples to oranges.

by gcanyon

3/16/2026 at 9:17:56 PM

Features like hardware FP8 support definitely make it apples-to-oranges.

by pdpi

3/16/2026 at 11:32:33 PM

But doesn't the Apple M series NPU support FP8, and as it's a monolithic die (except for the GPU in the M5 Pro and Max) it could be argued it has hardware FP8 support, no?

by philjohn

3/17/2026 at 1:52:12 AM

By that logic, on the M4 (which still has the GPU on the same die as the CPU), CPU cores have hardware accelerated raytracing, which is obviously nonsense.

by pdpi

3/17/2026 at 11:48:41 AM

Apple's hardware does not support FP8 (neither the ANE NPU, or the new "neural accelerator" tensor cores), though the most recent variant supports INT8.

by llm_nerd

3/16/2026 at 11:47:42 PM

I thought the M5 had FP16 support, and not FP8.

by badc0ffee

3/17/2026 at 8:14:19 AM

It doesn't matter, because you will never find M5 chips on cloud offerings, or server racks.

It is kind of rediculous that the only server option with Apple hardware has been to stack up mac minis.

They got rid of the server and workstation market, focusing on consumers only.

by pjmlp

3/16/2026 at 9:14:11 PM

Grace GB10, Vera's predecessor, had a single core performance comparable to M3 so I guess we can expect at least M4 level performance now.

by storus

3/16/2026 at 9:30:02 PM

Isn't the GB10 a Mediatek chip and not directly related to the Grace datacenter CPU?

by porphyra

3/16/2026 at 10:48:11 PM

The DGX Spark (and the white box variants of it) run on the Grace Blackwell GB10 "superchip".

by llm_nerd

3/16/2026 at 10:00:45 PM

More fair to say it's completely unrelated to the Grace data center CPU.

by wtallis

3/16/2026 at 9:12:36 PM

M5 are 9-18 cores and optimized for power-efficiency, those are more like Xeons, with 200-300W TDP, I'd bet.

by d_silin

3/16/2026 at 9:22:29 PM

If M5 has 9-18 cores and takes ~20w, then that's ~1-2w per CPU core. If these are 200-300W, and have ~100-200 CPU cores, then guess what? That's also ~1-2w per CPU core.

Xeons, Epycs, whatever this is - they are all also typically optimized for power efficiency. That's how they can fit so many CPU cores in 200-300W.

by kllrnohj

3/16/2026 at 8:31:53 PM

So does this cut out Intel/x86 from all the massive new datacenter buildouts entirely? They've already lost Apple as a customer and are not competitive in the consumer space. I don't see how they can realistically grow at all with x86.

by tencentshill

3/16/2026 at 8:47:37 PM

Even Apple hardware looks inexpensive compared to Nvidia's huge premium. And never mind the order backlog.

x86 and Apple already sell CPUs with integrated memory and high bandwidth interconnects. And I bet eventually Intel's beancounter board will wake up and allow engineering to make one, too.

But competition is good for the market.

by alecco

3/16/2026 at 11:17:43 PM

Even with those advantages, Apple can't even sell datacenter hardware to themselves: https://9to5mac.com/2026/03/02/some-apple-ai-servers-are-rep...

by bigyabai

3/16/2026 at 11:47:02 PM

"And as the initial crop of Apple Intelligence features hasn’t been used as much as Apple expected"

Nah, as so-called "analysts" expected. The no-effort crybabies deriding Apple for being "behind on AI" have turned out to be, shocker of shockers, wrong. Anyone who even put a few minutes of thought into Apple's business realized that it (and its customers) didn't stand to benefit much from "AI."

It's sad that Apple hurried to pander to these clowns, only to be derided further... and to encounter the appropriate apathy from customers, who were and are doing just fine without asinine "AI" gimmicks.

by MoonWalk

3/17/2026 at 12:18:32 AM

Apple wouldn't have built the server capacity if they thought it wouldn't be used. It's indeed their own analysis.

In any case, that article is also looking forward to next-gen models like the sparse Gemini model Google trained for Siri. Apple Silicon simply isn't powerful enough to compete for that inference.

by bigyabai

3/16/2026 at 9:15:16 PM

Apple went from a high-end PC to a low-end AI provider due to blocking Nvidia on their platform.

by storus

3/17/2026 at 1:23:34 PM

Worse, because since they no longer care about workstation market, there are pluggable cards, and no update to the chese grater, which the Studio is not comparable to.

They also dropped the ball on the data center, having left OS X Server behind.

Those markets are now served by Windows or Linux based configurations.

by pjmlp

3/16/2026 at 9:19:52 PM

>are not competitive in the consumer space

AFAIK they still dominate on clock rate, which I was surprised to see when doing some back of the envelope calculations regarding core counts.

I felt my 8 core i9 9900K was inadequate, so shopped around for something AMD, and IIRC the core multiplier of the chip I found was dominated by the clock rate multiplier so it’s possible that at full utilization my i9 is still towards the best I can get at the price.

Not sure if I’m the typical consumer in this case however.

by mikrl

3/16/2026 at 9:31:28 PM

Your 9900k at 5ghz does work slower than a Ryzen 9800X3D at 5ghz. A lot slower (1700 single core geekbench vs 3300, and just about any benchmark will tell the same story). Clock speed alone doesn't mean anything.

by kllrnohj

3/16/2026 at 10:13:28 PM

From the newegg listing:

>8 Cores and 16 processing threads, based on AMD "Zen 5" architecture

which is the same thread geometry as my 9900K.

My main concerns at the time were:

1. More cores for running large workloads on k8s since I had just upgraded to 128G RAM

2. More thread level parallelism for my C++ code

Naively I thought that, ceteris paribus and assuming good L1 cache utilization, having more physical cores with a higher clock rate would be the ticket for 2.

Does the 9800X3D have a wider pipeline or is it some other microarchitectural feature that makes it faster?

by mikrl

3/17/2026 at 1:03:24 AM

Comparing CPUs by clock speed doesn’t work. New CPUs are do more work per clock cycle.

A 9800X3D is twice as fast as your 9900K in benchmarks like GeekBench, despite having similar clock speed and the same core count.

If you could downclock the AMD part to 2.5GHz as an experiment it would still beat your 5GHz 9900K.

by Aurornis

3/16/2026 at 10:29:17 PM

You don't even need to go into the pipeline details. The 9800X3D has 8x more L2 cache, 6x more L3 cache, 2x the memory bandwidth than the now 8 years old i9 9900K. 3D V-cache is pretty cool.

by joefourier

3/16/2026 at 11:44:22 PM

I purposely picked a CPU with the same thread geometry as your 9900K to avoid calls of "apples & oranges" or whatever. If you want more threads, the 9950X is right there in the same socket. Or Core Ultra 9 285k. Either of which will run circles around a 9900K in code compilation.

You can research microarchitecture differences if you want, it's a fascinating world, or you can just skip to looking at benchmarks/reviews. Little hard to compare against quite that large of a generation gap, but eg https://gamersnexus.net/cpus/rip-intel-amd-ryzen-7-9800x3d-c... or https://www.phoronix.com/review/amd-ryzen-7-9800x3d-linux/2

by kllrnohj

3/17/2026 at 2:57:40 AM

The 9800X3D has wider everything. Decoder, execution ports, vectors, cache, memory bandwidth...

by zzzoom

3/17/2026 at 3:03:03 AM

I think my i9 was released right after the Spectre and Meltdown mitigations in 2019, but I seem to remember even more recent vulns in that family… so that could also be a factor.

by mikrl

3/16/2026 at 9:32:10 PM

A 9700X is twice the performance of a 9900K and M5 Max is almost 3X the performance. The megahertz myth is a myth.

by wmf

3/16/2026 at 10:15:30 PM

I replied to the sibling comment: I was making simplifying assumptions for two specific use cases and naively treated physical cores and clock rate as my variables.

by mikrl

3/17/2026 at 1:13:05 PM

But why? That's like trying to determine which car is faster by looking at only at the rpm.

by joefourier

3/17/2026 at 1:04:27 AM

Yes, but core count and clock speed of a nearly 10 year old CPU are meaningless when comparing to current processors.

by Aurornis

3/16/2026 at 10:19:08 PM

Ahhh, so is this a chip "more optimised" for connecting GPU's to reality ... or are they skipping the GPU step entirely? Are GPU's only for training now?

by RantyDave

3/16/2026 at 10:29:06 PM

have you seen this: https://chatjimmy.ai/

It's quite impressive what purpose build inference can/will do once everyone stops trying to become kind of the best model.

by cyanydeez

3/16/2026 at 10:59:23 PM

Wow impressive. What's the story with this?

by redwood

3/16/2026 at 11:06:03 PM

It's a tech demonstrator for a company that turns models into custom silicon for fast inference. In this case llama3.1-8b https://taalas.com/products/

by jffry

3/17/2026 at 2:24:07 AM

Is this an ASIC? Or FPGA? Or something even more exotic?

I’m guessing it’s some form of ASIC because I can’t imagine crafting the logic of Llama on silicon is a very quick or easy job. Not that doing it on an ASIC is a piece of cake either.

by gizajob

3/17/2026 at 1:26:37 PM

An ASIC is custom silicon, no?

Anyways, I found this article discussing it a bit more: https://www.eetimes.com/taalas-specializes-to-extremes-for-e...

"Taalas is borrowing some ideas from the structured ASICs of the early 2000s to make its hardwired model-specific chips. Structured ASICs used gate arrays and hardened IP blocks, changing only the interconnect layers to adapt the chip to a specific workload. At the time, this was seen as a more cost-effective alternative to a full-custom ASIC that was more performant than an FPGA."

"Taalas changes only two masks to customize a chip for a specific model, but the two masks can change both model weights and dataflow through the chip. On the HC1, the model and its weights are stored on the chip using a mask-ROM-based recall fabric paired with a (programmable) SRAM, which can be used to hold fine-tuned weights and/or the KV cache. Future generations of chips may split the SRAM onto a separate chip, meaning they could be denser than the HC1."

by jffry

3/16/2026 at 11:19:49 PM

Taalas hardware implementation of Llama 3.1 8B They claim 16k tok/s vs Cerbras at 2k. https://taalas.com/products/

by hmartin

3/16/2026 at 8:30:28 PM

> Purpose-Built for Agentic AI

From the "fridge purpose-built for storing only yellow tomatoes" and "car only built for people whose last name contains the letter W" series.

When can this insanity end? It is a completely normal garden-variety ARM SoC, it'll run Linux, same as every other ARM SoC does. It is as related to "Agentic $whatever" as your toaster is related to it

by dmitrygr

3/16/2026 at 9:15:03 PM

> It is as related to "Agentic $whatever" as your toaster is related to it

These things have hardware FP8 support, and a 1.8TB/s full mesh interconnect between CPUs and GPUs. We can argue about the "agentic" bit, but those are features that don't really matter for any workload other than AI.

by pdpi

3/16/2026 at 10:38:48 PM

The huge interconnect would also useful be for HPC tasks. The FP8 not so much, HPC still loves FP64.

by pezezin

3/16/2026 at 9:23:05 PM

mem bw between cores matters for .... literally all workloads that are not single-core (read: all). And FP8 matters not at all cause inference on cpu is too slow to be of any use whatsoever in the days of proper accelerators

by dmitrygr

3/16/2026 at 9:19:57 PM

Would cloud gaming platforms benefit from the interconnect?

by kibibu

3/16/2026 at 9:24:23 PM

Don't think they would. Games aren't nearly as hungry for memory bandwidth as LLMs are. Also, I expect that the VRAM/GPU/CPU balance would be completely out of whack. Something would be twiddling its thumbs waiting for the rest of the hardware.

by pdpi

3/16/2026 at 8:41:30 PM

The power and importance of marketing is deeply underappreciated by us technical types.

by dpe82

3/16/2026 at 9:17:11 PM

And yet more than a little Gavin Belson "Box III" vibes here. Fortunately, no signature edition.

by LogicFailsMe

3/16/2026 at 9:35:49 PM

I don’t underappreciate it, but I do despise it.

by dwb

3/16/2026 at 9:50:37 PM

> It is a completely normal garden-variety ARM SoC

To mis-quote the politician quip:

How can you tell a marketer is lying?

Answer: His/her mouth is moving.

by pwg

3/16/2026 at 9:41:13 PM

"democratize access to AI and accelerating innovation."

So they make inference cheaper and the models get even worse. Or Jensen Huang has AI psychosis. Or both.

Here is a new business idea for Nvidia: Give me $3000 in a circular deal which I will then spend on a graphics card.

by rka128

3/16/2026 at 10:13:53 PM

Me too plz. To quote (more or less) Harvey Pekar: “I’m trying to sell out, but nobody’s buying!”

by kwertyoowiyop

3/18/2026 at 3:07:45 PM

The most interesting part is that Nvidia intend to sell this CPU separately, meaning you dont need to buy Nvidia GPU to use it.

Other than Hyperscaler ARM has yet to enter the server market and it might well be Nvidia that makes a different.

by ksec

3/16/2026 at 8:44:26 PM

I'm assuming this is for tool call and orchestration. I didn't know we needed higher exploitable parallelism from the hardware, we had software bottlenecks (you're not running 10,000 agents concurrently or downstream tool calls)

Can someone explain what is Vera CPU doing that a traditional CPU doesn't?

by rishabhaiover

3/16/2026 at 9:18:35 PM

> you're not running 10,000 agents concurrently or downstream tool calls

Cursor seem to be doing exactly that though

by kibibu

3/16/2026 at 8:50:04 PM

Lots and lots of CPUs pooled. Faster more efficient power RAM accessible to both GPU and CPU. IIUC.

by urig

3/16/2026 at 9:07:17 PM

But at what stage are we asking for that RAM? if it's the inference stage then doesn't that belong to the GPU<>Memory which has nothing to do with the CPU?

I did see they have the unified CPU/GPU memory which may reduce the cost of host/kernel transactions especially now that we're probably lifting more and more memory with longer context tasks.

by rishabhaiover

3/16/2026 at 9:33:43 PM

Does this mean their gaming GPUs are becoming less in demand, and therefore cheaper/more available again?

by recvonline

3/17/2026 at 12:00:56 AM

Absolutely not, unfortunately.

The problem is not that gaming GPUs are in demand, it’s that selling silicon to AI center buildouts is so absurdly profitable right now they just aren’t making many gaming GPUs.

If you can only get so many mm^2 of dies from TSMC, might as well make 50x selling to AI providers.

by Teknoman117

3/17/2026 at 1:26:15 PM

Check the GTC 2026 agenda, there are hardly any graphics programming talks.

At least there are a few cool ones about programming CUDA directly in Python.

by pjmlp

3/18/2026 at 3:08:41 AM

Yep, it's a sad time for the gaming world.

by Teknoman117

3/16/2026 at 9:34:55 PM

It means it will be profitable to mine crypto again

by TheRoque

3/16/2026 at 9:44:51 PM

No.

by wmf

3/16/2026 at 9:21:03 PM

Am I crazy, or is Jensen's statement a copy-paste from ChatGPT?

(Could be both)

by kibibu

3/16/2026 at 9:23:33 PM

If AI is so great why should he not use it?

by wmf

3/16/2026 at 11:35:29 PM

Should work on building the AI Jensen. Maybe it's already the AI Jensen

by magackame

3/16/2026 at 9:11:02 PM

Are we rapidly careening towards a world where _only_ AI “computing” is possible?

Wanted to do general purpose stuff? Too bad, we watched the price of everything up, and then started producing only chips designed to run “ai” workloads.

Oh you wanted a local machine? Too bad, we priced you out, but you can rent time with an ai!

Feels like another ratchet on the “war on general purpose computing” but from a rather different direction.

by FridgeSeal

3/17/2026 at 4:28:36 AM

The World's First Central Sloppressing Unit

by simulator5g

3/17/2026 at 8:11:24 AM

what a bizarre title

by _s_a_m_

3/16/2026 at 10:42:31 PM

A GPU purpose-built for Slop.

by dude250711

3/17/2026 at 9:23:48 AM

[flagged]

by felixsells

3/17/2026 at 12:09:49 AM

[dead]

by gpubridge

3/16/2026 at 9:40:39 PM

[flagged]

by anesxvito

3/16/2026 at 9:43:34 PM

> Tauri over Electron is the same argument applied to desktop apps.

you lost me here but still got my upvote. Tauri and Electron are pretty much the same, compared to local-first vs cloud SaaS.

by brazukadev

3/16/2026 at 8:49:04 PM

What the heck is agentic inference and how is it supposed to be different from LLM inference? That's a rhetorical question. Screw marketing and screw hype.

by urig

3/16/2026 at 8:49:20 PM

Who wants general computing anyways?

by BoredPositron

3/16/2026 at 9:04:09 PM

China will beat this....

Seems like a triumph of hype over reality.

China can do breathless hype just as well as Nvidia.

by KnuthIsGod