Viewing profile — Zylokloto

Is there any particular reason why you respond like this? Do you think this adds anything to this discussion at all? Did you even take the time and effort to read my comment?

So first of, there is no proof of him having had sex with a minor/kid on that island. It might be a lot stronger of an evidence from others but i was not able to find anything. And…

Intereresting point I will try to see if i can find more about it. I would assume apple has a lot more magic sause than just TSMCs best node and a little bit of ARM architecture ad…

Yeah I know but that wasn't what I was hinting at. I mean you expect some company which did some android rom, a smartphone, IoT Robots stuff a car?! and now makes a chip which can …

I have 2 4090 and last time i played around with it, the context window killed it for me. The normal LLMStudio stuff works great, but then i tried out anything with subagent things…

There are sooo sooo many students and normal people using these devices or corporate laptops never seen inferencing or anything like this, its so hard to trash the storage of them,…

Ah sorry you are right I missread what you meant by it.

I'm just a german dude not finding this 'joke' funny. I'm not an apple fanboy nor a xiamoi fanboy. Its impressive that a random chinese company was able to catch up to the 2th rich…

I'm not a pro-china poster, i'm from germany and didn't find this 'funny'. Apple is the second richest company on the world (which doesn't need help/protection?!) and they have exp…

You will not feel a difference in coding/dev work between M1 to M6 in a relevant manner.

Last time the benchmark numbers from apple were stupid and they still are. I just wait for practical real live numbers with real stacks i use. And btw. they reduced the memory band…

You can already run it locally its just not the same. It is still slow, a lot slower than what you are used to with claude and co. And as soon as you increase context size, your me…

I do not find it funny tbh. I'm very very surprised that Xiaomi matches Apples speed even with the newest release, its not diminishing Xiaomis success.

GLM 5.3 in ~q4 quantiziation with 4bits

you can't link onchip memory.

In best case its also on-chip high speed ram.

To experiment, its still al ot cheaper to prepare everything locally and then just rent a GPU Node on all of these non hyperscalers. 1-2$ / hour. I'm not regretting my setup at hom…

Nvidia is quite open in publishing and talking about their models though.

That would be a complete defeat for a Frontierlab / Company like Google/Deepmind. They will never ever do like this. They for sure would add architecture ideas from other research/…

Another thing you should do. How do people even become like 60-70 or 80 around me without all this magic? tztztzt

I personally do not remember this time of Stack Overflow. I had some helpful people helping me on IRC / Quakenet. But the hugo example i found very interesting because it was the l…

As soon as the AI progress 'slows down' a little bit and its a little bit more clear what makes sense of converting into a hardware model, there will be for sure something. It eith…

Might just be preparation for their IPOs. But the agentic layer is being worked on, agents will start consuming more and more tokens

Yeah StackOverflow which was either telling you to use google or it was so specific, that no one wanted/could respond. Even a year ago when i was trying to do a hugo template manua…

We finetune LLMs. Small ones like Gemma 4 for semantic tasks. There are plenty of areas were we need people to do this for insurances, banks etc. AI/ML exists on many levels.