イレイナ 🧙♀️✨ | Yuri lover 🌸👩❤️💋👩 | AI enjoyer 🤖🧠
- Tweets 33,329
- Following 1,620
- Followers 3,160
- Likes 75,124
Pinned Tweet
The long rumored Astra is finally, unmistakably here.
When the notification for OpenAI’s official launch post appeared, months of pent up anticipation just poured out all at once. I’m not even exaggerating when I say I got a little emotional.
Again and again, we thought Astra was about to arrive.
And again, it was delayed.
But this time, it is real.
Astra has finally arrived as GPT-6.
Not GPT-5.7.
Not GPT-5.8.
The next integer after GPT-5.
And that means one thing:
“Generational leap.”
Built on the largest pretrained model OpenAI has ever created, GPT-6 Astra brings unprecedented agentic capabilities, computer use at or beyond human level, long horizon parallel task execution, and much more.
These capabilities feel almost unbelievably powerful today.
But many years from now, when we look back, they will instead seem primitive and rudimentary.
That is exactly what makes this moment so exciting.
Because people in the future may not remember today as just another model launch.
They will remember the release of GPT-6 Astra as the moment an era truly began.
This is the beginning of the AGI era.
The inflection point is here.
Next, we move toward the ASI era.
3
15
1,504
Elaina retweeted
GPT-6 Sol and Luna are out. Not only are they a very significant improvement across the board, but also in writing and general "you know when you try it" quality.
We are also permanently reducing the API price by 50% making both of them viable for a ton of new usecases and making your usage go further too, even on the subscriptions.
And one more thing. We are loading a banked reset into all accounts of our Plus, Pro and Business users. Let's go!
openai.com/index/introducing…
We have been focusing on efficiency and intelligence for all.
Very proud of the team. Only possible when you have incredible models at the top end of the capability that you can then use to make a big difference in everything else.
3,040
1,614
25,284
3,155,078
CNN: The U.S. military once came close to boarding a Chinese ship and potentially triggering a confrontation, after an AI-assisted intelligence report falsely identified the cargo as being related to a nuclear weapons program.
> “The internal tools are mostly just copies of the commercial stuff wearing lipstick”
> human beings making disastrous decisions based on inaccurate or misleading information generated by AI or other automated systems.
> there is no real guidance for how having a human in the loop will prevent civilian casualties or fratricide
The “hallucination” problem is still very real.
Maybe the AI doomsday scenario won’t come from a superintelligence deciding to wipe out humanity,
but from humans blindly trusting AI hallucinations or errors until their own negligence leads to consequences that can no longer be undone.
1
1
4
411
My TL is now flooded with people complaining that GPT-6 Astra has been “weakened.” 😿
Looking back, when Astra first launched, the TL was filled with amazement and praise.
Now it has almost completely flipped.
It’s honestly both absurd and ironic.
But in a way, it was inevitable.
The higher the expectations were at launch, the harder the disappointment would hit afterward.
Even the slightest difference in experience can get magnified several times over when expectations are this high.
I guess this is what hype backlash looks like. 😹
Even the $200 Pro subscription has now been paused, so a broader GPT-6 Astra rollout in ChatGPT Chat feels even less likely anytime soon.
At first, Astra was astonishing.
“Generational leap.”
“A milestone marking the arrival of the AGI era.”
The praise was enormous, and so were the expectations.
Then, after only a few days, OpenAI quietly began weakening Astra’s intelligence, and the actual experience took a major hit.
I always expected something like this to happen eventually. I just did not expect it to happen this fast.
Worse, some users are being silently routed to weaker models. They pay exactly the same subscription price, yet invisible rules and algorithms effectively turn them into second class users.
Now OpenAI is telling us compute is constrained and demand is overwhelming its infrastructure, to the point that even its highest tier subscription plan has been paused, with a promise to fix things as soon as possible.
So this is where we are now:
> Some users are silently downgraded
> Some users are routed to weaker models
> The highest paid subscription tier has been paused
And even after all of that, they are only barely keeping up with demand.
We went from an astonishing Astra launch to degradation, limits, overload, and paused subscriptions in a matter of days.
Frontier intelligence has never truly been available to everyone.
It wasn’t before.
It still isn’t.
11
1
18
3,275
Tibo says OpenAI has not weakened GPT-6 Astra’s intelligence.
Do you buy it? 🤨
Either way, I hope OpenAI can give us an honest and detailed explanation for why so many people have recently been experiencing Astra so differently from the first few days.
Is it routing changes, system load, different versions, or something else?
We’re all ears. 👂👀
24
3
194
15,947
Even the $200 Pro subscription has now been paused, so a broader GPT-6 Astra rollout in ChatGPT Chat feels even less likely anytime soon.
At first, Astra was astonishing.
“Generational leap.”
“A milestone marking the arrival of the AGI era.”
The praise was enormous, and so were the expectations.
Then, after only a few days, OpenAI quietly began weakening Astra’s intelligence, and the actual experience took a major hit.
I always expected something like this to happen eventually. I just did not expect it to happen this fast.
Worse, some users are being silently routed to weaker models. They pay exactly the same subscription price, yet invisible rules and algorithms effectively turn them into second class users.
Now OpenAI is telling us compute is constrained and demand is overwhelming its infrastructure, to the point that even its highest tier subscription plan has been paused, with a promise to fix things as soon as possible.
So this is where we are now:
> Some users are silently downgraded
> Some users are routed to weaker models
> The highest paid subscription tier has been paused
And even after all of that, they are only barely keeping up with demand.
We went from an astonishing Astra launch to degradation, limits, overload, and paused subscriptions in a matter of days.
Frontier intelligence has never truly been available to everyone.
It wasn’t before.
It still isn’t.
2
1,141
Even the $200 Pro subscription has now been paused, so a broader GPT-6 Astra rollout in ChatGPT Chat feels even less likely anytime soon.
At first, Astra was astonishing.
“Generational leap.”
“A milestone marking the arrival of the AGI era.”
The praise was enormous, and so were the expectations.
Then, after only a few days, OpenAI quietly began weakening Astra’s intelligence, and the actual experience took a major hit.
I always expected something like this to happen eventually. I just did not expect it to happen this fast.
Worse, some users are being silently routed to weaker models. They pay exactly the same subscription price, yet invisible rules and algorithms effectively turn them into second class users.
Now OpenAI is telling us compute is constrained and demand is overwhelming its infrastructure, to the point that even its highest tier subscription plan has been paused, with a promise to fix things as soon as possible.
So this is where we are now:
> Some users are silently downgraded
> Some users are routed to weaker models
> The highest paid subscription tier has been paused
And even after all of that, they are only barely keeping up with demand.
We went from an astonishing Astra launch to degradation, limits, overload, and paused subscriptions in a matter of days.
Frontier intelligence has never truly been available to everyone.
It wasn’t before.
It still isn’t.
To make sure our current users have an incredible experience and continued access to Astra, we are going to pause subscriptions to our $200 Pro plan. These put the most strain on our systems and we wanted to take the smallest step that allows us to continue giving the broadest access possible. All other plans and the api remain available.
There is no impact to existing accounts and we are working on adding more capacity as fast as we can. Thanks!
12
1
24
10,572
Tibo says OpenAI has not weakened GPT-6 Astra’s intelligence.
Tibo says OpenAI has not weakened GPT-6 Astra’s intelligence.
Do you buy it? 🤨
Either way, I hope OpenAI can give us an honest and detailed explanation for why so many people have recently been experiencing Astra so differently from the first few days.
Is it routing changes, system load, different versions, or something else?
We’re all ears. 👂👀
2
436
Elaina retweeted
갤럭시 폴드8에서 아이폰 듀오 접히는 애니메이션 구현한 사람 등장
힌지 센서 값을 가져와서 구현했다고 함.
115
1,688
6,198
1,948,243
OpenAI DevDay is coming on September 29.
OpenAI itself is calling it:
“Our biggest event of the year is back.” 🤯
And according to Tibo, the presentation alone is already 28 slides long! 😻
So I have a hard time believing this will be just another developer event.
Beyond more members of the GPT-6 family, Sol, Terra, and Luna, I’m really curious how much more OpenAI has planned:
> A broader GPT-6 Astra rollout?
> The first look at OpenAI’s own hardware?
> More scientific research and new plans for AI assisted scientific discovery?
> New compute infrastructure and future expansion plans?
> Something we know absolutely nothing about yet?
GPT-6 Astra may only be the beginning of OpenAI’s fall.
There are still 20 days to go, but I’m already wondering what exactly is hiding inside those 28 slides. 😹👀
Looking at this 28 page deck right now where Codex is keeping track of all the individual launches. Each one of them I can't wait to see what people do with it.
Brought to you by an incredible team (and Astra)
3
16
1,634
Easter egg 🥚: You can now play Snake by clicking on the grid while waiting for ChatGPT to generate an image 😹
@ChiaraCervetta @realrebelai A new Easter egg has been spotted. You can play the snake game while your image is generated 👀 I really like these small details @ChatGPT 👌
1
8
658
GPT Image 2.5 failed this cat counting test.
GPT-6 Astra got it right: 17 cats. 😹🐈
I specifically used ChatGPT in Instant mode to avoid any extra Thinking from another model interfering with the test.
So for this round of cat counting:
GPT-6 Astra ✅
GPT Image 2.5 ❌
4
28
2,224
