Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
DiabloD3
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
61.
▲
by
DiabloD3
3mo ago
All using ZFS, too
62.
▲
by
DiabloD3
3mo ago
I'm already playing 2.1, and have been since it came out like 3? 4? days ago. Seeya sometime next decade.
63.
▲
by
DiabloD3
3mo ago
As a reminder, the most dangerous job in Russia is "opposition leader".
64.
▲
by
DiabloD3
3mo ago
MTP models share internal state with the main model, and also refer to parameters in the model. They are more like a single model that has two separate attention head mechanisms.
65.
▲
by
DiabloD3
4mo ago
They weren't the first to do MTP like this, and arguably did it wrong: the MTP heads are kept in a separate file and have to be welded in by the inference engine. Qwen 3.6 shipped with working MTP first, and had working MTP in llama.cp
66.
▲
by
DiabloD3
4mo ago
Your father should only be looking at Kelly Blue Book values. What brand ends up retaining theirs year after year? Toyota. He's right to buy American made when it comes to cars, we know what we're doing... he's just gonna hav
67.
▲
by
DiabloD3
4mo ago
In a way, its funny you say Toyota is the foreign brand. Of domestically produced GM, Chrysler, or Ford vehicles, most are shipped to foreign customers; of domestically sold GM, Chrysler, or Ford vehicles, most are manufactured by foreign f
68.
▲
by
DiabloD3
4mo ago
No.
69.
▲
by
DiabloD3
4mo ago
If it wasn't for Linux and open source, I very much could!
70.
▲
by
DiabloD3
4mo ago
Thats neat.
71.
▲
by
DiabloD3
4mo ago
Ahh, I misread it.
72.
▲
by
DiabloD3
4mo ago
This is a weird way for Apple to admit the Mac is dead.
73.
▲
by
DiabloD3
4mo ago
So, by definition, you did edit it to change the typo.
74.
▲
by
DiabloD3
4mo ago
When I'm on Linux, I use KDE exclusively now (although I'm hoping Cosmic replaces it someday). The apps that use QML are fine even on decade old CPUs with absolute shit iGPUs, still better than even absolutely awfully designed Win
75.
▲
by
DiabloD3
4mo ago
They have their own LLM, might as well use it and then screw their biggest customers that are in the business of LLM-aaS
76.
▲
by
DiabloD3
4mo ago
This is a weird way for Nvidia to announce they're going out of business.
77.
▲
by
DiabloD3
4mo ago
I notice you say, specifically, Qt Widgets. Yes, classic Qt Widgets still doesn't allow for hardware acceleration. However, the majority of Qt UIs you deal with are in Qt Quick, which is hardware accelerated. Almost all of KDE is Qt Qu
78.
▲
by
DiabloD3
4mo ago
Windows is so fucking old that I think it has a right to try again. And, btw, the reason Microsoft even bothered is because (dun dun dunnnn) lots of internal apps at Microsoft were written with Qt, not MFC, and leadership got pissed when th
79.
▲
by
DiabloD3
4mo ago
It hasn't made any inroads on the desktop though... all anyone did was just package their own SWA into a self-contained browser that serves its own content. They continue to be websites, with all the pitfalls of them. I don't need
80.
▲
by
DiabloD3
4mo ago
If the NSA director said this, literally and verbatim, then him, and maybe several of his predecessors, should be remembered as the worst directors the NSA has ever had. LLMs cannot create anything new, they can only repeat their training d
81.
▲
by
DiabloD3
4mo ago
I don't get the point of this. The world is trying to make computers faster and more accessible, more web UI slop isn't going to help that. Dumping Javascript entirely is the first step on that road.
82.
▲
by
DiabloD3
4mo ago
I wonder if this is a technique that's been passed down through the Armstrong family for generations
83.
▲
by
DiabloD3
4mo ago
Oh, there is. It was invented by China in the 10th century.
84.
▲
by
DiabloD3
4mo ago
Thats a weird way to announce the NHL is going out of business.
85.
▲
by
DiabloD3
4mo ago
I suspect this person didn't read the README or the `--help` at all. For example, `--cpu-moe` makes it not offload MoE layers to the GPU, which drops performance about a quarter, but only keeps the dense and important layers on the GPU
86.
▲
by
DiabloD3
4mo ago
Not being glib at all. They're overpriced toys. Too little CPU, too little GPU, too little RAM, and what RAM you do have has too little memory bandwidth to serve both the CPU and GPU at the same time (which still hasn't been fixed
87.
▲
by
DiabloD3
4mo ago
But why would developers _step down_ to a Mac?
88.
▲
by
DiabloD3
4mo ago
Interesting.
89.
▲
by
DiabloD3
4mo ago
Okay, please show it to me. All LLMs can do is statistically reproduce their training material; they cannot originate new ideas, they cannot reason about ideas, they cannot reason about themselves, and they have no concept of self. If there
90.
▲
Jeff Geerling: "Maybe don't rip off other people's video content"
(xcancel.com)
12 points
by
DiabloD3
4mo ago
|
0 comments
More ›