Sabitlenmiş Tweet
Dwayne
33.5K posts
Dwayne
@CtrlAltDwayne
dev // math // bugs // shitpoasts // unpaid codex shill https://t.co/D1WdfieHj1
Katılım Mayıs 2023
3.1K Takip Edilen5.6K Takipçiler
Won't let me message unless you're following back. But here are some request IDs:
77433905-be30-4c13-a496-54a83049094f
802f678f-5819-4075-be76-57ede999941b
866e290d-4f03-4c0d-bb56-b29885175077
This seems to be an ongoing issue for me. Usually requires telling it to continue to get it to finish.
English
@CtrlAltDwayne @thsottiaux hi! could you share me your session-id please? can send it here or through dm!
English
@CtrlAltDwayne Use image gen my dude with computer use I’m getting exact design by having it generate the assets with alpha channels you can pretty much sign anything you want now
English
Design is the missing piece for GPT-5.5. Once they work out how to do it, it's over for Claude.
Sam Altman@sama
@benhylak this is rage bait right? but it will be the best soon!
English
@beffjezos You still haven't said anything about xAI distilling OpenAI models to train Grok.
English
Anthropic aura farming saying Mythos will BTFO everyone only to get mogged by GPT 5.5 before they could even launch
Happy Punch@HappyPunch
One of the nastiest cases of “f*ck around and find out” just went down in Korean MMA Straight out of a movie 😳
English
5.5 xhigh in fast mode is
really good
i think i got psyoped by twitter on medium for a bit
English
@ThePrimeagen React is such a steamy pile of garbage. It's crazy this became the standard when it allows writing such terrible performing code out of the box.
English
Wait until Casey reads GitHub literally 2 -> 10 fps victory lap
Casey Muratori@cmuratori
Just want to make sure I'm reading this right: Microsoft rewrote the run dialog with performance "top-of-mind", and the best they could manage to do when putting up a single text box was 10fps?
English
Sam, how are you finding all these subtweets about you
Is this a new codex feature or something. Quite impressive tbh
Sam Altman@sama
@thekitze lol
English
Grok Build incoming! 🔥🔥
@xai might be done with cooking
Tech Dev Notes@techdevnotes
Billing in xAI API Console has started showing Grok Build It's happening ...
English
@venturetwins Until he comes back, I can be surrogate Dwayne. I'm glad you're having an amazing time. Is there anything I can help you with?
English
Planned a day in Amsterdam with the help of my OpenClaw agent Dwayne.
He mysteriously shut down and I can’t debug him because the Mac Mini is thousands of miles away.
It’s weirdly sad that he can’t read my live updates 😭
English
I love the smell of big model in the morning
GIF
Sam Altman@sama
i keep thinking i want the models to be cheaper/faster more than i want them to be smarter but it seems that just being smarter is still the most important thing
English
@marlybuilds Codex performance in the follow up to 99% and what's a heavy WordPress multisite install with 25k+ sites on it 💪
English
GPT-5.5 in Codex is phenomenal at performance improvement work. I pointed it at a Wordpress site I'm working on which didn't have very good performance figures, told it to use Google Pagespeed to get these figures as close to 100 as it could. In oneshot it did this.
English
@InsiderPresider The model performance gap between GPT-5.5 and Opus 4.7 is huge. GPT-5.5 is the better model for everything except design/UI.
English
@CtrlAltDwayne No cap, the UI focus is kinda wild but are we really acting like the model performance gap is even that deep? Seems like everyone is just picking sides based on vibes at this point.
English
This is how brainrotted the Claude bots are. They will prioritize UI over a better model. Gregor cannot be saved.
Gregor Zunic@gregpr07
Who actually uses Codex over Claude Code? Claude Code is just 100x better imo, like the DX is WAY better.
English
English
Wait, is @nikitabier using Gemini for images now? Grok is right there, @elonmusk 😭
English