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?
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.
> 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.
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.
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.
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.
The difference is the algorithm needed to be embedded in an asic is static for Bitcoin. Models are changing too fast
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.
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.
But why wouldn't nvidia be expected to be a major successful player in that development?
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.
Yup. Pretty much. Any model can be a a hard wired circuit.
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:
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
[dead]
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.