[jroesch]

Note: this is old work, and much of the team working on TVM, and MLC were from OctoAI and we have all recently joined NVIDIA.

[sebmellen]

Is there no hope for AMD anymore? After George Hotz/Tinygrad gave up on AMD I feel there’s no realistic chance of using their chips to break the CUDA dominance.

[comex]

Maybe from Modular (the company Chris Lattner is working for). In this recent announcement they said they had achieved competitive ML performance… on NVIDIA GPUs, but with their own custom stack completely replacing CUDA. And they’re targeting AMD next.

https://www.modular.com/blog/introducing-max-24-6-a-gpu-nati...

[dismalaf]

IMO the hope shouldn't be that AMD specifically wins, rather it's best for consumers that hardware becomes commoditized and prices come down.

And that's what's happening, slowly anyway. Google, Apple and Amazon all have their own AI chips, Intel has Gaudi, AMD had their thing, and the software is at least working on more than just Nvidia. Which is a win. Even if it's not perfect. I'm personally hoping that everyone piles in on a standard like SYCL.

[steeve]

We (ZML) have AMD MI300X working just fine, in fact, faster than H100

[latchkey]

https://x.com/dylan522p/status/1871287937268383867

[quotemstr]

The world is bigger than AMD and Nvidia. Plenty of interesting new AI-tuned non-GPU accelerators coming online.

[fweimer]

Isn't AMD rather strong in the HPC space?

Quite frankly, I have difficulty reconciling a lot of comments here with that, and my own experience as an AMD GPU user (although not for compute, and not on Windows).

[llm_trw]

Not really.

AMD is constitutionally incapable of shipping anything but mid range hardware that requires no innovation.

The only reason why they are doing so well in CPUs right now is that Intel has basically destroyed itself without any outside help.