Carl

There was an issue right when I hit my usage limits. I tried to use one of my banked resets and now neither has one applied but they all are gone. @thsottiaux how can that happen?!

@maria_rcks As mentioned in the Discord, you both cooked with those 😍🔥

New icons from Julius and me

@maria_rcks @theo t3 code is a "wrapper" around a harness

@maria_rcks @convex Maria's humor is on another level. AND I LOVE IT😂

@convex how i felt typing this

Migrates your project over to @convex

@theo looks great! can't wait to have it in t3 code

Now we're getting somewhere

recommended reading. good advise for now.

@sama I provably didn't make the coolest thing but I made a Discord Server Transcript Tool which lets you use an API point with a Bot Token and Channel ID and creates a beautiful transcript with features like Download as html, pdf, or JSON. Not the coolest but I'm proud that I made it!

i'd love to see interesting things people have built with 5.6 sol. i will send the person who made the coolest thing a special gift from the openai archives.

Morning. The last 48 hours of Codex and ChatGPT Work have been intense! Three important updates: - Temporarily removing the 5 hour usage limit restriction for all Plus, Business and Pro plans - Rolling out changes that will make GPT 5.6 Sol more efficient across the board and that will be reflected in less usage being used so that it can take you further. Exact impact to be quantified and shared - We hit 6M active users, and are landing a usage reset in the next hour Go do things

@maria_rcks Congrats Maria, you made it. Your mom must be proud of you!

HOLY SHIT THANK YALL (& theo) SO MUCH

@maria_rcks Remember me before your fame!

i just woke up to 9k followers 😭 what the fuck theo tysm

@theo This is actually helpful. Theo, you really are the best! Please never stop making Videos ever!

@maria_rcks I have to live with a Plus sub. I hope you can feel my pain...

I've already burned through three 5h limits, 😔

fwiw, I think that turning Codex into ChatGPT Desktop is a generational fumble

We audited SWE-Bench Pro, one of the most widely used AI coding benchmarks, and found it no longer reliably measures frontier coding capability. We find 30% of SWE-Bench Pro tasks to be broken, and are retracting our previous recommendation that the research community use it as a leading coding eval. openai.com/index/separati…

Agreed.

@MarcosHernanz I'd agree with 5.6 Sol Ultra and 5.6 Luna Comparisons. 5.6 Sol without Ultra is way better than Opus 4.8 and 5.6 Terra is basically gpt 5.5 but cheaper (on paper). Based on OpenAI's Benchmarks, 5.6 Terra is using more Tokens while being same level as 5.5 making it a tie.

My guess: GPT 5.6 Sol Ultra = Fable 5, but 2x cheaper GPT 5.6 Sol = Opus 4.8, but 2x cheaper GPT 5.6 Terra = Sonnet 5, but 2x cheaper GPT 5.6 Luna = Haiku 4.5, but 2x cheaper Basically: GPT = Claude, but 2x cheaper

The best proof that Fable and GPT-5.6 are “next generation” is the surge of other labs catching up to the previous generation

@FindLogan @OpenAI I wouldn't quite say that. Fable has more general knowledge due to its more parameters, while 5.6 Sol is at around the same parameter level as 5.5

@OpenAI GPT Sol will be the Fable 5 killer. Where do we place bets?

GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday. We’re expanding preview access globally now.

@onlyskypace Bottom and Top side are 9m; left and right side are 8m. 8*9=72. Are y'all high or am I dumb?