dust42

7 ms·

61.
by dust42 8mo ago
KV caching means that when you have 10k prompt, all follow up questions return immediately - this is standard with all inference engines. Now if you are not happy with the last answer, you maybe want to simply regenerate it or change your l
62.
by dust42 8mo ago
Unfortunately Qwen3-next is not well supported on Apple silicon, it seems the Qwen team doesn't really care about Apple. On M1 64GB Q4KM on llama.cpp gives only 20Tok/s while on MLX it is more than twice as fast. However, MLX has
63.
by dust42 8mo ago
From personal experience I'd like to add the last 5% take 95% of the time - at least if you are working on a make over of an old legacy system.
64.
by dust42 9mo ago
Indeed. Rather than having the company telling me that they did great I'd rather make up my own mind and watch the video.
65.
by dust42 9mo ago
> People writing and maintaining software need to optimize for simplicity, readibility, maintainability. Whether they use an LLM to achieve that is seconday. The humans in the loop must understand what's going on. In a perfect world
66.
by dust42 9mo ago
Linux is nowadays mostly sponsored by big corporations. They have different goals and different ways to do things. Probably the first 10 years Linux was driven by enthusiasts and therefore it was a lean system. Something like systemd is typ
67.
by dust42 9mo ago
> AIs have endless grit (or at least as endless as your budget). That is the only thing he doesn't address: the money it costs to run the AI. If you let the agents loose, they easily burn north of 100M tokens per hour. Now at $25&#x
68.
by dust42 9mo ago
Max was always closed.
69.
by dust42 9mo ago
Prepare to go back to newsgroups with NNTP / UUCP. With today's uSD cards that should create a pretty decent, offline store and forward national discussion platform. No programming needed, it's all there already, just forgott
70.
by dust42 9mo ago
Scroll down on the Huggingface page, there are code examples and also a link to github: https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-Base
71.
by dust42 9mo ago
> This stuff is relatively new, I don't think anyone has truly figured out how to best approach LLM assisted development yet. Exactly. But as you say, there are so many people riding the hype wave that it is difficult to come to a s
72.
by dust42 9mo ago
> You've then proceeded to make a whole host of other assumptions about me rather than engage with any of the points. I beg your pardon! You said Europeans are at war currently -please don't deny it, it is up there- and being m
73.
by dust42 9mo ago
> for only the truly incorruptible would colloborate with Nazis As I'd assume from your comments you consider yourself truly incorruptible. So does that mean collaboration with the Nazis is acceptable for you? Which actually would b
74.
by dust42 9mo ago
Well, the EU was quick to copy this. The Swiss Jaques Baud was slapped with the same measures. I saw only one youtube video of him and in my memory he was well outspoken and considerate - but definitely not mainstream but also not a conspir
75.
by dust42 9mo ago
There are reports from passengers that the train rattled before the accident. So my guess is a broken wheel rim and subsequently the train derailed at the track switch then also damaging the opposite track. Accident location: https:/&
76.
by dust42 9mo ago
From the aerial imagery it looks like the accident sequence started at the track switch [1]. The RENFE is rested south of it and the Iryo is north. Quite similar to the 1998 Eschede ICE accident which started with a broken wheel rim and fin
77.
by dust42 9mo ago
Personally, I'd prefer if the AI models would start with a proof of their own statements. Time and again, SOTA frontier models told me: "Now you have 100% correct code ready for production in enterprise quality." Then I run i
78.
by dust42 9mo ago
30 years ago I attended a university lecture in an economics class and the professor spoke about the economic consequences of global warming - some places will be better off and plenty of places will be worse off. There will be water shorta
79.
by dust42 9mo ago
They are just fully AI driven on every level but also there are people listening to it. Youtube currently gets swamped with AI generated content - music is only one part of it. For example there are now endless history documentaries. To get
80.
by dust42 9mo ago
Yes, apart from voice cloning nothing really new. Kokoro is out since a long time and it supports at least a few languages other than english. Also there is supertonic TTS and there is Soprano TTS. The latter is developed by a single guy wh
81.
by dust42 9mo ago
Good quality but unfortunately it is single language English only.
82.
by dust42 9mo ago
> This is going to sound sarcastic, but I mean this fully: why haven't they merged that PR. I would go even further, why have they not created at least one less complex project that is working and ready to be checked out ? To me it
83.
by dust42 9mo ago
I don't get all this frontier stuff. Up to today the best model for coding was DeepSeek-V3-0324. The newer models are getting worse and worse trying to cater for an ever larger audience. Already the absolute suckage of emoticons sprink
84.
by dust42 9mo ago
I used 700W per H200 = 11.2 per 16 GPUs. I didn't include CPU and rest of the rack. So yours is a better approximation. One has to keep in mind that the benchmark that was done is synthetic. This makes sense because it makes it reprodu
85.
by dust42 9mo ago
If I followed the links correctly this benchmark was made on a 16xH200. At current prices I'd assume that is a system price of around $750,000. The year has 86400*365 = 31536000 seconds. Thus 63072000000 tokens can be generated. As pri
86.
by dust42 9mo ago
From my experience AI is only good with mainstream coding tasks. Javascript, node, react, crud. Whatever it has seen in overabundance, it is good with. Already with typescript it is less strong than with javascript. It is just a clever toke
87.
by dust42 9mo ago
Distilling from a closed model like GPT-4 via API would be architecturally crippled. You’re restricted to output logits only, with no access to attention patterns, intermediate activations, or layer-wise representations which are needed for
88.
by dust42 9mo ago
That's right, they all give a variant of that, for example Qwen says: I am Qwen, a large-scale language model developed by Alibaba Cloud's Tongyi Lab. Now given that Deepseek, Qwen and Kimi are open source models while GPT-5 is no
89.
by dust42 9mo ago
So if you subtract linux and LLVM and Webkit and Java, what is left of Google? Absolutely nothing. Well, a mostly empty, dysfunctional mono repo lacking the main dependencies.
90.
by dust42 9mo ago
> Google has poured untold millions into open source over the last couple of decades, not just by sponsorships, but also by employing contributors, etc. And Google has profited untold hundreds of billions of open source over the last cou

More ›