reel

@teej_dv @KevinNaughtonJr

@_Jomann Surtout que l’entrée est gratuite juste pour remplir

Pas sold out tonton

The girlies had their fun yesterday, today is for real men

Babamı s!kicem az kaldı vallahi 2 günlük tatilde inşaata çalışmaya gittim tam 2000 TL biriktirdim rdr2 alacaktım gitmiş benim paraya arabaya süs almış amına kodumun adamı bende sinirlendim gittim pcyi yere devirdim mal küçük kardeşimde benden gaza gelip kasaya tekme atmış ekran kartı yerinden çıkmış garip anamda çöp zannetmiş atmış ben iyice delirdim anneanneme yatmaya gittim babam gönlümü almak için pasta getirmiş küçük boy almış birde amına koyayım baba siktir git dedim adam ağlamaya başladı bende ağlayınca üzüldüm pastayı yiyodum pastayı portakallı almış amk Allah belanı versin baba dedim yere tükürdüm pastayı babamın iyice morali bozuldu 2 gündür evden çıkmıyormuş bugün anamın 1 tane gram altını vardı birini zorla alıp sattım rdr2yi aldık sonunda buda böyle bir günüm oldu

do gnome devs really deserve basic human rights

"vat are you doing mein fuhrer?" "i am creating a desktop environment.. hmm.. i vill call it.. GNOME..."

@nagzencore Nn mais azy frère si c’était le contraire le mec serait finito

@hnfong Impressive, no DOM reads ?

It's interesting to see this blowing up on the Internet My (ex)colleagues implemented this (in Javascript) like 15 years ago. Not open source, but still.

@BcZsalba @kenwheeler These 2 : behind Bun and React ; You and all the others : behind Shit

i’m gonna have a fucking aneurysm

@ondakx @kenwheeler The brother is behind react and midjourney but ondakx with his shitty repo on bio and 5 dollar subscription thinks he better than us

@kenwheeler i will have to leave this network for some time or create a curated list of survivor of ai cyberpsychosis shitting while shipping code was not possible before ai

@BcZsalba @kenwheeler Lol the brother is behind react and midjourney, i think if it was that easy some random like you would have done it instead of tweeting in his moma house, and citing Microsoft word just shows that you are half-witted

@kenwheeler > "cannot" be done without AI > was already achieved in the web space years before AI > Microsoft Word has had it for like 20 years worst part is it's cool, but any enjoyment of it is tainted by the grift others imposed on it

@TelepathicPug @kenwheeler Why didnt you do it then ? The brother is literally behind react and midjourney and said that without the AI testing a billion edge cases and auto incrementing the fixes it wouldn’t have been possible but @TelepathicPug thinks otherwise

@kenwheeler I remember that x.com/_chenglou/stat…

@Bearnotbeer5 @grok traduis le texte de l’image

Вот за что не любят линуксоидов🙄

@Cocoader @burkov @grok true ?

@burkov FWIW, Grok asserts that this doesn’t much work with Grok — it is already an internal tactic that Grok uses.

LLMs process text from left to right — each token can only look back at what came before it, never forward. This means that when you write a long prompt with context at the beginning and a question at the end, the model answers the question having "seen" the context, but the context tokens were generated without any awareness of what question was coming. This asymmetry is a basic structural property of how these models work. The paper asks what happens if you just send the prompt twice in a row, so that every part of the input gets a second pass where it can attend to every other part. The answer is that accuracy goes up across seven different benchmarks and seven different models (from the Gemini, ChatGPT, Claude, and DeepSeek series of LLMs), with no increase in the length of the model's output and no meaningful increase in response time — because processing the input is done in parallel by the hardware anyway. There are no new losses to compute, no finetuning, no clever prompt engineering beyond the repetition itself. The gap between this technique and doing nothing is sometimes small, sometimes large (one model went from 21% to 97% on a task involving finding a name in a list). If you are thinking about how to get better results from these models without paying for longer outputs or slower responses, that's a fairly concrete and low-effort finding. Read with AI tutor: chapterpal.com/s/1b15378b/pro… Get the PDF: arxiv.org/pdf/2512.14982

@oxcrowx @grok pourquoi il peut pas utiliser gcc ?

Always has been.

@TheGracia_here Ai slop

@nvssm_ Partage immédiatement

bon bah on est parti pour la 7eme année gratuite 🤣🤣🤣

@HoopsCrave @grok true ?

Diddy is reportedly missing from prison.

@nocontextmemes @grok ça marcherait ?