@JustinBleuel @ChatGPTapp I don’t see all of my. Codex sessions sometimes until I switch to chronological order than back to projects
English
js
2.8K posts
js
@jothecoder
i code stuff | master prompter or whatever
Katılım Ağustos 2014
806 Takip Edilen599 Takipçiler
Every bug in @ChatGPTapp is getting fixed
With the help of codex (and the rest of the lovely team and their codexes) along with a 7pm iced americano there will be zero bugs
This is a formal request for tiny nits, error states, broken ui, etc
The tinier the better!
Justin@JustinBleuel
@theOpusLABS @Fixlation7 @ChatGPTapp Yeah I do the same. Thanks again for flagging! fix for this is up and will go out soon 🙏
English
@theo Oh yeah, and my laptop won't run Codex App because it's too old. I'm not buying a new laptop just to run it. Terminal windows work fine. I use Textmate to view the code in those very rare cases where I find that to be helpful.
English
Are you still using the CLI versions of your preferred agent instead of desktop apps like Codex App, Conductor, or T3 Code?
Tell me why below. Genuinely curious.
Theo - t3.gg@theo
Just learned it's literally impossible to paste images into Claude Code over SSH. How do you CLI people live like this??
English
@yacineMTB benchmarks are all talk, show the model doing some real work without failing instead
English
Lmaooo yoooo .... Wordddd???? This is a Copilot Billing Preview for the usage-based billing.
I hope this estimate is $2646.63 off from the actual price bc.....😂
I think we're going to start seeing the "Let's just write this in assembly to ultra optimize it to the tiniest size possible" being the move
English
Want to (officially) use Codex at work?
Send this post to your CTO to bring your team to Codex. Eligible enterprise customers who switch in the next 30 days get 2 free months of Codex usage for new users.
English
Doom Swift is in good progress, it's now fully playable and almost 100% accurate. Missing the music and a few other things but Codex /goal did it, in almost full autonomy. 100% Swift and software rendered.
English
@yacineMTB no way you’re actually reviewing all of the code like you claim to do
English
It worked :)
The inference for this model in particular went from 22ms to 16ms
kache@yacineMTB
/goal optimize this code, use `date` to check the time and don't stop until I wake up at 7 am
English
@kevinkern my own profile, i want the ai to pretend to be me, i hate it when i get a dumb claude instance that refuses to fill out ach / wire info. i use chrome-cdp-skill to do all my payroll related stuff, qbo related stuff, my landlord rent payments for my bakeries, etc
English
has anybody found a way to perma allow remote debugging in chrome? it stops my long runs all the time.
English
English
@JeffKazzee @GeoffreyHuntley i think none = automatic reasoning? i forget tho
English
Recently I had moved Junior (our slackbot) to GPT 5.4 models. I figured performance was similar to Anthropic, I'd try it out. Especially since they have more capabilities in some areas.
Today I moved it back to Anthropic models. The models just werent performing that well. They - just like you probably see in Codex - require a lot of supervision to get them to do what you ask. I spent a bit of time but couldn't easily work past the issues.
We went from typically gpt-5.4 medium thinking back to sonnet and the results are night and day.
The bot once again behaves how we expect it to.
English
@yacineMTB did you apply to the party? Im on hour 11 of reverse engineering and its used 40% weekly limit on a pro plan
English
It’s not attendees it’s all applicants
vogel@ryanvogel
yooooo let go all gpt 5.5 attendees get 10x rate limits massive W @OpenAI
English
Every ripped person i see now i just assume they’re on Reta 😂
English
In my experience GPT, Gemini, and DeepSeek v4 are strongest in theoretical physics. This benchmark suggests the same.
CritPt evaluates language models on solving unpublished, frontier-level physics problems that require genuine research-scale reasoning. The benchmark comprises 71 challenges (70 test challenges and one example), created by over 50 active physics researchers across 30 institutions and spanning 11 physics subfields
English
@ThePrimeagen I really like to vibe code
Scrolling X, podcasts, deeply in the flow of idea marination
English
I really just like to program
Hands on keyboard, music, deeply thinking and enjoying the process
English
@jothecoder used someones coupon code on here that I trust, paid a premium tho and will use diff source next time
English