zargon

3 ms·

1.
by zargon 2mo ago
I promise I searched on my phone before posting that, but it looks like I misinterpreted an hf search result that was actually Qwen-Image-2512 but the url was truncated to Qwen-Image-2 due to my screen width (ugh). For some reason I was thi
2.
by zargon 2mo ago
It would make more sense to compare with Qwen Image 2, since that was the last open weights Qwen model. Edit: This is wrong.
3.
by zargon 2mo ago
Memory allocated for HBM is memory taken away from DDR production.
4.
by zargon 2mo ago
"Your Brain Is Not an Onion With a Tiny Reptile Inside" is a really stupid and pretentious article arguing against a straw man.
5.
by zargon 2mo ago
I have a tendency to use documentation as a quality bar. Byre might be great but I can't continue reading this slop.
6.
by zargon 2mo ago
Fun for about 5 minutes but screeches to a halt when ships start to teleport. Only takes a few minutes to be able to get to wave 12 and then there's no progress after that.
7.
by zargon 2mo ago
I didn't think to check for mods. Thanks.
8.
by zargon 2mo ago
This perspective that consciousness cannot be programmed can only make sense if you're a dualist. We don't know how consciousness arises. If you're a naturalist it can't be ruled out based on the simplicity of the algori
9.
by zargon 2mo ago
Unpopular opinion, but for me, no amount of puzzles/story can offset the utter obnoxious frustration of the time loop.
10.
by zargon 2mo ago
> I've been using AI increasingly more and more > genuinely We can tell.
11.
by zargon 3mo ago
Thanks for this suggestion. I have used thumb trackballs before and they cause me some strain. But the Skylla looks very promising.
12.
by zargon 3mo ago
The Glove80 looks close to what I want, but that thumb cluster looks basically as bad as the Ergodox one. Too many keys overall, especially pinky ones.
13.
by zargon 3mo ago
AGPLv3 is awesome, great choice!
14.
by zargon 3mo ago
Aris Mpitziopoulos at cybenetics.com
15.
by zargon 3mo ago
I'm not sure whether this comment is serious or not. Poe's law strikes again.
16.
by zargon 3mo ago
> Do not give yourself diabetes by eating vast amounts of sugar. If you are doing this, stop. Not having diabetes decreases CVD risk. Not eating vast quantities of sugar is good advice, but it's not the direct cause of T2 diabetes.
17.
by zargon 3mo ago
Yes, llama.cpp does that.
18.
by zargon 3mo ago
KL divergence is nothing close to a replacement for benchmarks. As flawed as benchmarks are, KL divergence is a barely useful signal. The fact that Unsloth only just started publishing KL divergences shows how unserious the quantization spa
19.
by zargon 3mo ago
As a replacement for ollama, LM Studio is probably the most appropriate. Though I haven't tried Unsloth Studio yet. You get more control using llama.cpp directly, but there's more of a learning curve. vLLM has the best performance
20.
by zargon 3mo ago
> ollama's model index typically only distinguishes variants by their size Don't use ollama. The entire project is just a series of stupid decisions like this.
21.
by zargon 3mo ago
I have hearing loss and movie theaters are deafening.
22.
by zargon 3mo ago
https://xkcd.com/605/
23.
by zargon 3mo ago
V4 Flash fits entirely in two RTX Pro 6000s without any quantization at all.
24.
by zargon 3mo ago
> The point of local models is privacy, offline use, and maybe no guard rails. But also, control and consistency. A local model cannot be changed out under your feet like an API model can be.
25.
by zargon 4mo ago
> emerging practice of using Q4 quants and Q8 KV cache for local inference That's not an emerging practice, it's a tested strategy that is these days only used as a last resort by those desperate to fit a model in memory. Some
26.
by zargon 4mo ago
I tried to evaluate Jellyfin recently but dropped it pretty quick. Live (OTA) television is barely a thought (seems to be acknowledged that no dev uses it), and you have to figure out a source for guide data. I have complaints with Plex&#x2
27.
by zargon 4mo ago
Probably not.
28.
by zargon 4mo ago
The name is abliterate. It's a specific method of ablation.
29.
by zargon 4mo ago
If you bought it from one of those ebay outfits that specialize in memory modules, they also often offer a lifetime warranty. Usually those sellers put their own sticker on the module. Worth a check.
30.
by zargon 4mo ago
> However, if you buy the CPU yourself, neither Intel nor AMD provides any guarantee that the CPU will work with overclocked memory. There's no guarantee that it works with standard clocked memory either.

More ›