Darkhex
378 posts
The potato is right
Gpt 5.5 has gotten dumber a lot
Gpt 5.6 is most likely releasing tomorrow
So if i were you , i wouldn't touch my usage until then
Best case we get a reset , worst case you have used most of your usage on lobotimzed 5.5 instead of maximizing on 5.6
🥔🥔🥔@argofowl
gpt 5.5 hasn't been the same the past 3 days it's making unbearable mistakes on all reasoning efforts i'm so tired of it
English
@TheRohanVarma @LexnLin Its so crazy you guys Arent releasing this to pro users.
English
@LexnLin All of our models, harness, and tooling come together nicely with Sites to enable creating very useful and well designed software!
English
5.5 Medium made this? No way
I hope Sites has some crazy magic behind it that the generated UI really looks that good
OpenAI@OpenAI
Building apps has never been easier. With Sites, Codex can turn your work, ideas, and plans into an interactive website or app your team can explore, use, and share with a URL. Rolling out to Business and Enterprise plans, before expanding more broadly.
English
OpenAI has seriously messed up Codex $100 plan limits.
It feels like the old $20 plan now.
The 5 hour usage just keeps depleting even on a short response.
Claude $100 max is giving way more value right now. It’s so hard to finish 5 hour limits there.
@OpenAI @thsottiaux fix
English
@TappyLeaks69 grok cli can spawn 100 subagents at once and properly coordinate them and now my grok heavy membership has a model that i enjoy using. This is a great update
English
This is huge. Grok build is a very good cli with alot of potential
I did a full RE of the harness & it had some of the highest potential I've seen
Grok really contending now to be your daily driver 👀
xAI@xai
Composer 2.5 is now available inside Grok Build. Composer 2.5 is a fast, highly intelligent model that excels on long-running tasks and following complex instructions.
English
Back to Cursor; been months since I have used it
Everyone around on the feed are praising Composer 2.5, I want to give it a try too
Rudrank Riyam@rudrank
Codex is so bad today I need to setup my workflow on Claude too so I can switch easily I am at a point where I wasting more time on iterations on design and copy vs just using Claude instead
English
English
@Timirin11 Idk I just feel like you didn’t have to post it here… I was praying snap took it down already… 🙏🏾 in hindsight, I think she’s having an episode and I don’t really want to indulge her.
English
Codex $100 plan after being back to 5x is just...
You can finish the weekly limit in a day (without using /fast)
English
@rezoundous 30% of weekly limits just in one day of light work
English
@clockworkhr_ @rezoundous Bro ive used goals for 15 hours and im at 60% weekly. $100 plan is a JOKE rn
English
@rezoundous I beg to differ. My codex ran on a task last night for 21 hours. Not even halfway past my usage. For Claude, it's a different story.
English
No more 2x for ChatGPT Pro Lite plans
And 25x -> 20x for Pro plans 🥲
It's going to be rough in Codex
Kappaemme@Kappaemme1926
Codex 2x usage limits end today. If you think the limits are bad now, June is about to be painful. I’m on the $100 plan and even with 2x limits, I ran out in 2 days after the reset. Do you think OpenAI will extend the promo, or are we just cooked?
English
@thsottiaux Tomorrow morning? Thats literally when our limits reset anyway
English
Five million users would agree. Resetting the limits tomorrow morning to celebrate.
Time to go /fast
Siqi Chen@blader
nothing like switching to claude for a few days to try out a new model and going back to codex xhigh to remind you how much better 5.5 is right now it's really not close
English
What is happening today
Riley Brown@rileybrown
I hope you know tomorrow is going to be a VERY big day in the world of AI agents...
English
GOOD NEWS
Pro $200 still has the 20x offer after 31.05 :)
Slopware Engineer@slopwareindy
@xdotdan @kr0der Here it is in black and white (in red) 😛 I forgot the only thing that changes for 200 ppl it's the 25x 5 hour limit which is literally if zero consequence. I've never gotten under 70% with normal usage. Unless they don't respect this...
English
@thsottiaux Before you go zen or whatever pls fix the rate limits issue, also give us back gpt 5.5 from release, it feels like you replaced it with gpt 3.1 while the limits are exhausting as if its gpt 6.9
English