Re:Making AMD GPUs competitive for LLM inference
by jsheard on Monday December 23, 2024 @08:38PM
(#42498998)
Tinygrad was another one, but they ended up getting frustrated with AMD and semi-pivoted to Nvidia.
[ Parent
| on HN ]
Re:Making AMD GPUs competitive for LLM inference
by nomel on Monday December 23, 2024 @10:29PM
(#42499521)
This is discussed in the lex Friedman episode. AMD’s own demo would kernel panic when run in a loop [1].
Re:Making AMD GPUs competitive for LLM inference
by kranke155 on Tuesday December 24, 2024 @04:19AM
(#42500712)
Interesting. I wonder if focusing on GPUs and CPUs is something that requires two companies instead of one, whether the concentration of resources just leads to one arm of your company being much better than the other.
Re:Making AMD GPUs competitive for LLM inference
by halJordan on Tuesday December 24, 2024 @08:07PM
(#42506147)
Nvidia maintains a competitive cpu...
Re:Making AMD GPUs competitive for LLM inference
by kranke155 on Wednesday December 25, 2024 @06:29AM
(#42508282)
I had no idea. Thanks for sharing.
Re:Making AMD GPUs competitive for LLM inference
by noch on Tuesday December 24, 2024 @12:52AM
(#42499975)
> Tinygrad was another one, but they ended up getting frustrated with AMD and semi-pivoted to Nvidia.
From their announcement on 20241219[^0]:
"We are the only company to get AMD on MLPerf, and we have a completely custom driver that's 50x simpler than the stock one. A bit shocked by how little AMD cared, but we'll take the trillions instead of them."
From 20241211[^1]:
"We gave up and soon tinygrad will depend on 0 AMD code except what's required by code signing.
We did this for the 7900XTX (tinybox red). If AMD was thinking strategically, they'd be begging us to take some free MI300s to add support for it."
---