Re:Making AMD GPUs competitive for LLM inference
by latchkey on Thursday August 10, 2023 @10:31AM
(#37076631)
> This is completely false.
It isn't completely false. You're doing super limited stuff as a hobbyist that barely works.
Re:Making AMD GPUs competitive for LLM inference
by superkuh on Friday August 11, 2023 @12:37AM
(#37085161)
The parent article is entirely about running and benchmarking Llama2-7B/13B. This is the "super limited stuff as a hobbyist that barely works" and I've run them at entirely usable speeds on the AMD RX 580. Just admit that you're wrong, or at least that you didn't actually read the article.
Re:Making AMD GPUs competitive for LLM inference
by latchkey on Friday August 11, 2023 @01:39PM
(#37091487)
"entirely usable" is not the same as "roi efficient"
> from your ignorant perspective
no need for the ad hominem.
Re:Making AMD GPUs competitive for LLM inference
by superkuh on Monday August 14, 2023 @02:05PM
(#37124369)
Ignorance is not an insult. It just became obvious that you were talking about a different concept (commercial use with big models) than the article itself and everyone else were talking about (7B/13B models). So I generously assumed you just hadn't read it (ignorance). I guess now that you've ignored that and doubled down I can assume you were/are just arguing in bad faith.
Now it's your turn Mr. "You're not going to find rx580's with enough vram for AI. Typically 4-8gb." This is completely false. Rather than acknowledging that you then tried to move the goalposts (much like I did in that past thread saying, "Oh, but maybe it's just my region where they don't.") It looks like we both behave a bit silly when trying to save face when we're wrong.