all of which one notes typically run on three distinct sets of chips (Nvidia for GPT-5, Amazon Trainium for Anthropic and Google TPUs for Gemini)
I was previously under the impression that ~all AI models ran on Nvidia, and that was (probably) a big part of why Nvidia now has the largest market cap in the world. If only one out of three of the biggest models uses Nvidia, that’s a massive bear signal relative to what I believed two minutes ago. And it looks like the market isn’t pricing this in at all, unless I’m missing something.
(I assume I’m missing something because markets are usually pretty good at pricing things in.)
xAI and Meta still use Nvidia. Almost every non-frontier-lab, non-Chinese AI chip consumer uses Nvidia
And Alphabet, Amazon, and Broadcom, the companies that design TPU and Trainium, have the 4th, 5th, and 7th biggest market caps in the world.
I think it’s possible that the market is underpricing how big a deal Anthropic and Google DeepMind, and other frontier labs that might follow in their footsteps, are for overall AI chip demand. But it’s not super obvious.
Google TPUs have been competitive with Nvidia GPUs for years, but Google did not sell its TPUs, instead only renting them out via GCP, until very recently as it is now starting to actually sell them too.
Other GPUs and custom silicon like Trainium are used for inference these days, but training is almost exclusively done with Nvidia GPUs and Google TPUs. It’s pretty easy to take a trained model and make it possible to do inference with it using different chips, as we see for example with open-weight models being used on Apple silicon, and DeepSeek models being served on Huawei Ascend GPUs.
I still expect the majority of inference being done today to be on Nvidia GPUs, a notable portion on TPUs, and then some non-negligible amount on other chips. (I haven’t actually estimated this though.) Very roughly, I think in 1-3 years Nvidia will have a lot of competition for inference compute, though maybe not that much competition for training compute apart from TPUs, since CUDA is somewhat of a hard-to-overcome moat.
Google DeepMind uses Nvidia very sparingly if at all. AlphaFold 3 was trained using A100s but that’s the only recent use of Nvidia by GDM I’ve heard of. I think Google proper, outside GDM, primarily uses TPUs over GPUs for internal workloads, but I’m less sure about that.
Google does buy a lot of Nvidia chips for its cloud division, to rent out to other companies.
That shouldn’t matter too much for stock price, right? If Google is currently rolling out its own TPUs then the long-term expectation should be that Nvidia won’t be making significant revenue off of Google’s AI.
This caught me off guard:
I was previously under the impression that ~all AI models ran on Nvidia, and that was (probably) a big part of why Nvidia now has the largest market cap in the world. If only one out of three of the biggest models uses Nvidia, that’s a massive bear signal relative to what I believed two minutes ago. And it looks like the market isn’t pricing this in at all, unless I’m missing something.
(I assume I’m missing something because markets are usually pretty good at pricing things in.)
xAI and Meta still use Nvidia. Almost every non-frontier-lab, non-Chinese AI chip consumer uses Nvidia
And Alphabet, Amazon, and Broadcom, the companies that design TPU and Trainium, have the 4th, 5th, and 7th biggest market caps in the world.
I think it’s possible that the market is underpricing how big a deal Anthropic and Google DeepMind, and other frontier labs that might follow in their footsteps, are for overall AI chip demand. But it’s not super obvious.
Google TPUs have been competitive with Nvidia GPUs for years, but Google did not sell its TPUs, instead only renting them out via GCP, until very recently as it is now starting to actually sell them too.
Other GPUs and custom silicon like Trainium are used for inference these days, but training is almost exclusively done with Nvidia GPUs and Google TPUs. It’s pretty easy to take a trained model and make it possible to do inference with it using different chips, as we see for example with open-weight models being used on Apple silicon, and DeepSeek models being served on Huawei Ascend GPUs.
I still expect the majority of inference being done today to be on Nvidia GPUs, a notable portion on TPUs, and then some non-negligible amount on other chips. (I haven’t actually estimated this though.) Very roughly, I think in 1-3 years Nvidia will have a lot of competition for inference compute, though maybe not that much competition for training compute apart from TPUs, since CUDA is somewhat of a hard-to-overcome moat.
Both Google and Anthropic still use a lot of Nvidia compute; e.g. Trainium isn’t ready yet according to semianalysis.
Google DeepMind uses Nvidia very sparingly if at all. AlphaFold 3 was trained using A100s but that’s the only recent use of Nvidia by GDM I’ve heard of. I think Google proper, outside GDM, primarily uses TPUs over GPUs for internal workloads, but I’m less sure about that.
Google does buy a lot of Nvidia chips for its cloud division, to rent out to other companies.
That shouldn’t matter too much for stock price, right? If Google is currently rolling out its own TPUs then the long-term expectation should be that Nvidia won’t be making significant revenue off of Google’s AI.