HackerTrans
TopNewTrendsCommentsPastAskShowJobs

moelf

1,870 karmajoined 11 anni fa
https://github.com/Moelf/

Submissions

Discord Incident – Resolved

discordstatus.com
89 points·by moelf·2 mesi fa·59 comments

Jane Street Signs $6B AI Cloud Agreement with CoreWeave

coreweave.com
2 points·by moelf·3 mesi fa·0 comments

Vkdt – GPU raw photography workflow

github.com
2 points·by moelf·7 mesi fa·0 comments

OpenAI releases GPT-5.1 alongside eight new ChatGPT personality styles

arstechnica.com
4 points·by moelf·8 mesi fa·1 comments

[untitled]

1 points·by moelf·8 mesi fa·0 comments

It seems like the AI crawlers learned how to solve the Anubis challenges

social.anoxinon.de
177 points·by moelf·11 mesi fa·148 comments

comments

moelf
·5 giorni fa·discuss
frontier to laptop runnable open weight so far seems to be ~2 years latency, so maybe there's some hope
moelf
·16 giorni fa·discuss
even the submitted page doesn't, huge margin if you zoom out on mobile (Firefox)
moelf
·18 giorni fa·discuss
the original post says:

    So what now? The short answer is that I have (a) a nice couch; (b) a big TV; (c) a Steam Deck that I never use. I even have a dock for the Deck.
moelf
·3 mesi fa·discuss
how does Opus see the output? via HTML (which Typst can output) or visual tokens?
moelf
·4 mesi fa·discuss
any comparison with https://actualbudget.org/ ?
moelf
·6 mesi fa·discuss
what do people use for Neovim to integrate these models for tab-completion level of stuff. (i.e. non agentic/vibe coding)
moelf
·6 mesi fa·discuss
the most efficient languages are pretty unpopular, so this argument makes them even more efficient in reality?...
moelf
·6 mesi fa·discuss
>We worked out a special arrangement so that this server is physically held by a long time contributor with a proven track record of securely hosting services.

Not clear if "contributor" is a person or an entity. The "hosting services" part make it sound more like a company rather than a natural person.
moelf
·7 mesi fa·discuss
and HexOS I guess.
moelf
·7 mesi fa·discuss
you can always make a new vector that's orthogonal to all the ones currently used and see if the inclusion improves performance on your tasks
moelf
·7 mesi fa·discuss
>instead of spending massive compute rediscovering it from scratch every time

it's interesting that this paper was discovered by JHU, not some groups from OAI/Google/Apple, considering that the latter probably have spent 1000x more resource on "rediscovering"
moelf
·7 mesi fa·discuss
https://github.com/stevengj/subsuper-proposal
moelf
·7 mesi fa·discuss
https://www.androidauthority.com/graphene-os-major-android-o... hope fully soon
moelf
·8 mesi fa·discuss
(2018)
moelf
·8 mesi fa·discuss
the note is written in Typst!
moelf
·8 mesi fa·discuss
"Luckily," the hardware won't allow for 4k@120Hz on visually cutting edge games anyway.
moelf
·8 mesi fa·discuss
yeah it's just a Linux x86 desktop (Arch Linux) -- although, you'd likely want to make sure Jellyfin's hardware acceleration works well with AMD APU (last time I checked the AMD was under experimental)
moelf
·8 mesi fa·discuss
I guess right now the GPU is too weak. And ofc even if the hardware steps up, there are always root-kit games gatekeeping :(
moelf
·8 mesi fa·discuss
>Semi-custom AMD Zen 4 6C / 12T

the desktop is also x86, the VR headset (Frame) is ARM
moelf
·8 mesi fa·discuss
related https://news.ycombinator.com/item?id=45766138