logoalt Hacker News

coreylane • yesterday at 4:48 PM • 22 replies • view on HN

Remember when GPUs were the dominant tech to mine crypto, then ASICs and other spepcialized purpose-built hardware came out and made GPUs irrelevant for mining, wont that also happpen in regards to ai training/inference. like aws is attemppting to do with their trainium/inferentia chips. Sure they suck now, but won't they get better and thus lose the dependency on GPUs?


Replies

notnullorvoid • yesterday at 5:00 PM

It's already happened hasn't it. Google's TPUs are apparently much more efficient at inference workloads. However the hardware isn't available to the public like Nvidia chips are.

I'm unsure about the energy efficiency of Tenstorrent cards, but they are at the moment much more cost effective than Nvidia. But software support is far behind, I suspect if it wasn't the cost would be more equal.

➕ show 10 replies
lizknope • yesterday at 10:39 PM

All of the hyperscale data center companies have their own AI chips and their own CPUs as well. They don't want to be dependent on Nvidia (or AMD or Intel)

They still buy Nvidia and x86 chips but this is what the WSJ article from last week was saying. The hyperscalers are moving away and Nvidia is committed to TSMC wafer orders so they are funding smaller companies to buy Nvidia chips in exchange for equity.

Microsoft

https://blogs.microsoft.com/blog/2026/01/26/maia-200-the-ai-...

https://techcommunity.microsoft.com/blog/azureinfrastructure...

Google

https://cloud.google.com/tpu

https://cloud.google.com/products/axion

Amazon

https://en.wikipedia.org/wiki/AWS_Graviton

https://aws.amazon.com/ai/machine-learning/inferentia/

https://aws.amazon.com/ai/machine-learning/trainium/

davesque • yesterday at 11:08 PM

Apples and oranges, really. SHA-256 is a fixed function with no memory. So it lends itself to being hardwired into pure logic. A CPU or GPU was absurdly wasteful for it. That's why ASICs saw like 100x gains over general purpose hardware. Most of the repeatable tasks in AI workloads, like matrix and vector math at reduced bit widths, have already been encoded into hardware in modern GPUs.

bryanlarsen • yesterday at 4:50 PM

> Sure they suck now,

Some suck. Google's Tensor is an example of one that doesn't, that some consider superior to Nvidia's offering. There are likely others.

KeplerBoy • yesterday at 5:15 PM

To be fair GPUs have already changed a lot in response to the AI demands. A lot of silicon these days are tensor cores which have very few applications outside of deep learning.

__MatrixMan__ • yesterday at 5:55 PM

Inference, yes, see https://taalas.com/ and https://chatjimmy.ai/ As expected, efficiency boost is extreme. But you've got to pick the model you want to enshrine in silicon months or maybe years before you actually get that silicon (and maybe there are additional difficulties if the model is large. Maybe in a few years everybody will just want to use the same four or five models, with no tectonic shifts on the horizon. Then it'll be a no-brainer to make ASICs for those models.

Training? I'm not so sure. You need something flexible so that it can be trained. Burning the weights into the board is the opposite of that.

➕ show 2 replies
jack_pp • yesterday at 4:57 PM

The difference is the algorithm needed to be embedded in an asic is static for Bitcoin. Models are changing too fast

➕ show 1 reply
throwaway63467 • yesterday at 5:49 PM

NVIDIA GPUs stopped being GPUs a while ago, that didn’t happen for crypto, but now it seems the will drop the entire customer graphics market long term they just haven’t committed themselves to it. They can’t go back to GPUs anyway there are not enough gamers to reach the revenue numbers they need now and soon gaming/graphics revenue will be irrelevant for them. Most US companies won’t maintain a boutique of products that don’t have synergy or are actively competing for internal resources long term they put all their resources where the margins and growth are best and for NVIDIA that’s not gaming anymore.

➕ show 2 replies
logtempo • yesterday at 9:45 PM

https://spectrum.ieee.org/inference-hardware-revolution

Relevent article from september 26. To me it’s obvious gpu are a « first response » solution. But I think nvidia can sell a lot before AI R&D slow down (because of the demand or the tech, infra, energy… or whatever reason it will be)

ChainOfFools • yesterday at 5:30 PM

There are also the opposing architectural directions to extend toward:

one in the "Cerebras" direction of ever-larger scale general purpose HPC silicon running inference workloads. Cerebras may have skipped ahead to the last act of that scaling story, notwithstanding any further wafer-diameter enlargements past 300mm. Whether the ending is happy or not is still unclear. Lot of iterative room to grow into it further and they have a formidable patent moat waiting for conventional die-sized AI semi vendors when they inevitably run up against limitations of single-die designs.

the other is in the "Taalas" direction of freezing a specific model into something like high speed ROM / FPGA and getting orders of magnitude inference speed improvements - for that specific model. Like Cerebras, Taalas may also have jumped directly to the architectural telos of that philosophy, with only iterative improvements in cost per work unit left to grind, though like CBRS quite a lot of grinding left to do especially in getting cost per custom tapeout down to near-disposable (hopefully, recyclable) levels.

➕ show 1 reply
sanderjd • yesterday at 4:50 PM

But why wouldn't nvidia be expected to be a major successful player in that development?

➕ show 2 replies
m4rtink • yesterday at 6:23 PM

So will people buy gaming GPUs fron Nvidia again after how if abused them this time ? Or will the AI obsession be their downfall & example for others what not to do and how not to alienate the customers who get you where you are now?

TheAlchemist • yesterday at 8:34 PM

Yeah, that's my thesis since the beginning of this madness. I see it as a temporary blip (albeit it's already much much longer than I would expect !) and once the bubble pops, Nvidia will be back to their normal cruising speed. Which is a good company, making great products, but their valuation will be 1/10 of what it is now. And if the boom continues, there is no way either that their clients won't develop a product of their own. It just doesn't make sense to buy chips by the 10s of billions $ on which Nvidia makes >50% margin.

hermitShell • yesterday at 4:56 PM

This is why I think their pivot/venture to general robotics was smart. However, I doesn't seem like they will become profitable on that front because they were outcompeted pretty hard in most measures I can conjure.

FrustratedMonky • yesterday at 5:34 PM

Yes. Half dozen companies are designing special chips to bypass/replace the GPU.

Bitcoin was lucky for Nvida, then AI was lucky. Third times the charm?

On other hand, luck favors the prepared. Beyond luck, they did make their chips accessible, easier to program than other GPU's. So bitcoin miners and then AI could utilize them.

There are still a lot of people that want to run regular liner algebra.

➕ show 2 replies
IAmGraydon • yesterday at 5:32 PM

Mining ASICs are bound to a single cryptographic hashing algorithm, meaning they do one thing and they do it very fast. Something similar may make sense for running models locally, but I believe the more multifaceted Nvidia chips make more sense for training.

➕ show 1 reply
cmrdporcupine • yesterday at 8:55 PM

If you look at the sessions from 2026's "Hot Chips" there's like a dozen competitors and possibilities and new architectures and the like, yes. Some very compelling. And OpenAI and Google and others are also making their own chips.

The thing is, it's only NVIDIA really that has a proven product in high volumes pumping through TSMC with the existing software stack that the community has already proven. That's a huge difference. Everything else is riskier if you're building out in volume. For now.

IAmGraydon • yesterday at 5:35 PM

Check out Taalas, recently acquired by AMD. They printed a model directly to silicon and it generates 16,000 tokens per second. You can even try a demo here:

https://chatjimmy.ai/

➕ show 2 replies
itsthecourier • yesterday at 5:02 PM

won't

those Asics have a fixed algorithm

Cerebras already does that an need to make a new chip for each new model

that's why you couldn't make an ASIC for Ethereum

➕ show 3 replies
hparadiz • yesterday at 4:51 PM

Yup. Pretty much. Any model can be a a hard wired circuit.

quater321 • yesterday at 5:07 PM

[dead]