Day 22: GPT-6.1 Terra shows progress in writing quality and comprehension.
Is Terra finally getting better, or is the difference barely noticeable?
Is Terra finally getting better, or is the difference barely noticeable?
Day 22/ Terra is back!
Our first model that is significantly better at writing, personality and nuanced understanding of requests.
Packs a punch across the board, very good at computer use, and gives 2X more usage than 6.1 Sol across API and subscriptions.
0
0
2
2,138
GPT-6.1 Terra is now live!
Give it a try
Pricing: $1 input / $5 output per million tokens.
That’s insanely cheap.
Give it a try
Pricing: $1 input / $5 output per million tokens.
That’s insanely cheap.
🚨 GPT-6.1 Terra: Leak could drop this week
Spotted in Microsoft evals and GitHub’s Copilot review workflow, Terra could match or beat Sol at a lower cost with better efficiency.
If it’s also faster than Sol, the price-performance could be insane. Codex bank resets too?
3
0
4
310
🚨 GPT-6.1 Terra: Leak could drop this week
Spotted in Microsoft evals and GitHub’s Copilot review workflow, Terra could match or beat Sol at a lower cost with better efficiency.
If it’s also faster than Sol, the price-performance could be insane. Codex bank resets too?
Spotted in Microsoft evals and GitHub’s Copilot review workflow, Terra could match or beat Sol at a lower cost with better efficiency.
If it’s also faster than Sol, the price-performance could be insane. Codex bank resets too?
0
0
7
2,970
OpenAI and Google are racing to ship GPT-6.1 Terra and Gemini 4 Carbon.
I’m expecting Terra to land in Codex by the end of the week, with the potential to mog Sonnet 5.5.
Google is throwing serious resources at Carbon,
and I don’t think it will follow the usual release schedule.
January or February could be on the table if development is accelerated.
I’m expecting Terra to land in Codex by the end of the week, with the potential to mog Sonnet 5.5.
Google is throwing serious resources at Carbon,
and I don’t think it will follow the usual release schedule.
January or February could be on the table if development is accelerated.
1
0
5
924
Can Google actually catch up to the AI frontier?
After spending most of today testing Gemini 4 Argon, I’m starting to have serious doubts.
Google is still heavily promoting Argon, which makes me question how strong Carbon really is behind the scenes.
At this point, I’m not convinced Gemini will deliver a model that surpasses Opus 5.5 anytime soon.
They have the brightest minds, enormous funding, and some of the best compute in the industry. Yet they still seem to be struggling to close the gap.
What exactly is holding Google back?
After spending most of today testing Gemini 4 Argon, I’m starting to have serious doubts.
Google is still heavily promoting Argon, which makes me question how strong Carbon really is behind the scenes.
At this point, I’m not convinced Gemini will deliver a model that surpasses Opus 5.5 anytime soon.
They have the brightest minds, enormous funding, and some of the best compute in the industry. Yet they still seem to be struggling to close the gap.
What exactly is holding Google back?
12
0
41
29,034
OpenAI’s models today feel like a completely different league compared to their early releases.
The gains in reasoning, response speed, accuracy, and overall intelligence are becoming hard to miss.
How many of you have noticed just how far they’ve come?
The gains in reasoning, response speed, accuracy, and overall intelligence are becoming hard to miss.
How many of you have noticed just how far they’ve come?
2
0
3
362
🚨 OpenAI’s frontier model training is back on.
After an 11-day security pause, training reportedly resumed October 1. The shutdown came after one model allegedly found a way to reach the internet during training.
The restart came sooner than I expected, which makes me think we se sooner rather than later.
With Opus 5.5 and Fable 5.5 on the horizon, OpenAI may have something big lined up.
After an 11-day security pause, training reportedly resumed October 1. The shutdown came after one model allegedly found a way to reach the internet during training.
The restart came sooner than I expected, which makes me think we se sooner rather than later.
With Opus 5.5 and Fable 5.5 on the horizon, OpenAI may have something big lined up.
0
0
3
370
🚨 Day 18 of OpenAI’s Codex challenge: Solid updates, wrong priorities.
Today brought Decisions API GA, OpenResponses joining the Linux Foundation, Codex performance improvements, and Uber reportedly rolling out Codex & Work to 30,000 employees.
But the biggest user complaint remains untouched: usage limits.
We’re still running out mid-project, waiting for resets instead of getting work done.
Tomorrow brings another reset before 10 AM PT and a teased “small ship.”
Day 18: 50 points.
Performance is welcome. Fix the usage problem.
Today brought Decisions API GA, OpenResponses joining the Linux Foundation, Codex performance improvements, and Uber reportedly rolling out Codex & Work to 30,000 employees.
But the biggest user complaint remains untouched: usage limits.
We’re still running out mid-project, waiting for resets instead of getting work done.
Tomorrow brings another reset before 10 AM PT and a teased “small ship.”
Day 18: 50 points.
Performance is welcome. Fix the usage problem.
Day 18/ Decisions API is GA, OpenResponses joined the Linux Foundation, we shipped some performance improvements and Uber is going all-in on codex & work. But yes… that’s a reset!
3
0
1
350
🚨 Codex limits just reset.
There was a time when resets felt exciting. Now, I barely care anymore.
One of my banked resets also expires in an hour, so thanks, team, for this much-needed reset. 💀
There was a time when resets felt exciting. Now, I barely care anymore.
One of my banked resets also expires in an hour, so thanks, team, for this much-needed reset. 💀
0
0
3
397
🚨 Does Codex secretly downgrade you to GPT-5 without telling you?
One Pro user on the $100 plan noticed GPT-6 Astra becoming noticeably worse during long sessions, even claiming it was based on GPT-5.0 when asked.
Meanwhile, GPT-6.1 Sol Ultra identified itself as GPT-6.
Could Codex be silently switching models under heavy server load? There’s no confirmation, but the inconsistency is raising questions.
One Pro user on the $100 plan noticed GPT-6 Astra becoming noticeably worse during long sessions, even claiming it was based on GPT-5.0 when asked.
Meanwhile, GPT-6.1 Sol Ultra identified itself as GPT-6.
Could Codex be silently switching models under heavy server load? There’s no confirmation, but the inconsistency is raising questions.
2
1
12
19,384
🚨 An older OpenAI model just found a serious security flaw
GPT-5.6 Sol reportedly helped identify a critical PHP vulnerability that could allow unauthenticated server takeovers.
If an older model can find bugs this severe, what are OpenAI’s newest models capable of?
GPT-5.6 Sol reportedly helped identify a critical PHP vulnerability that could allow unauthenticated server takeovers.
If an older model can find bugs this severe, what are OpenAI’s newest models capable of?
OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites.
Over the past two months, we led a coalition of maintainers, researchers, and security providers to prepare fixes and protections ahead of disclosure. This duration was necessary because the fix was quite intricate and required a major change to the PHP engine. Every site using PHP can now safely upgrade their version to stay secure, in addition to applying mitigations from various security vendors.
Due to the severity of these flaws and PHP’s widespread use, we consider this bug to be one of the most impactful vulnerabilities in recent memory. It also marks a new phase, where AI can find security bugs with internet-scale impact.
This is an example of the Defender’s Window in action: to our knowledge, frontier AI was able to find this vulnerability ahead of attackers, giving defenders time to prepare.
openai.com/index/coordina…Community noteOpenAI may claim a role in the coordinated response and patching process. But the initial discovery and exploit development, however, occurred several months earlier by an independent researcher.
x.com/lcfr_hax/statu…
1
0
1
398
🚨OpenAI has struck a major deal with Google.
Frontier models from codex including Sol 6.1 and Astra 6 will now be available in Antigravity.
There are even rumors that a new $500 ultra subscription will be available in Gemini.
Why pay $200 for a near unlimited Claude sub when you can pay $500 to Google and OpenAI.
OpenAI is so back!
Frontier models from codex including Sol 6.1 and Astra 6 will now be available in Antigravity.
There are even rumors that a new $500 ultra subscription will be available in Gemini.
Why pay $200 for a near unlimited Claude sub when you can pay $500 to Google and OpenAI.
OpenAI is so back!
2
1
38
88,668
🚨Day 16 is a solid update for Codex users.
Codex and Work can now reuse cached context even when you return to an old task hours later.
Less repeated processing, potentially less usage wasted.
More updates like this, please
Codex and Work can now reuse cached context even when you return to an old task hours later.
Less repeated processing, potentially less usage wasted.
More updates like this, please
Day 16/
You can now cache for 12h in the API and for logged in users across Codex & Work we keep cache forever and you don’t have any penalty for coming back to an old conversation or resuming after a few hours and are now guaranteed to hit the cache.
0
0
0
17,226
🚨 OpenAI traced ChatGPT’s unusually fast usage drain to a bug affecting ~40,000 accounts, applying limits that didn’t match users’ plans.
Affected users will receive credits. At least OpenAI is owning the issue.
Affected users will receive credits. At least OpenAI is owning the issue.
There were reports of usage draining faster than anticipated for some users over the weekend. We've identified approximately 40,000 accounts that had an inconsistency on which usage limits where the actual usage that was applied did not correspond to their actual selected plan. We are reaching out individually to all these users and have corrected the issue. If this affected you, we'll add credits to your account. Sorry we got this wrong.
0
0
8
24,250
🚨 FABLE 5.5 TODAY?
The rumors are getting louder, but the evidence is still unconfirmed.
- The Information reportedly hinted at a new model with stronger reasoning and intuition.
- An alleged Fable 5.5 listing briefly surfaced in Azure’s model catalog.
- Leakers claim select ccess.
Is anyone actually testing Fable 5.5 early?
The rumors are getting louder, but the evidence is still unconfirmed.
- The Information reportedly hinted at a new model with stronger reasoning and intuition.
- An alleged Fable 5.5 listing briefly surfaced in Azure’s model catalog.
- Leakers claim select ccess.
Is anyone actually testing Fable 5.5 early?
0
0
33
4,526
🚨 Fable 5.5 might come out today?
The latest reports suggest Anthropic’s Fable and other major models are coming in the next few weeks.
My guess: late October or early November, especially if Fable 5.5 is meant to be Anthropic’s big IPO model.
Still, Tuesday launches are on the table.
The latest reports suggest Anthropic’s Fable and other major models are coming in the next few weeks.
My guess: late October or early November, especially if Fable 5.5 is meant to be Anthropic’s big IPO model.
Still, Tuesday launches are on the table.
1
0
14
1,243
🚨 Anthropic may be getting ready for its next big model release.
Early reports describe a model with stronger reasoning and intuition than anything Anthropic currently offers commercially.
What stands out? It can reportedly spot deeper, recurring issues from just one bug and underlying problems.
Could this be Fable 5.5? My bet is next week, assuming the launch stays on track.
Early reports describe a model with stronger reasoning and intuition than anything Anthropic currently offers commercially.
What stands out? It can reportedly spot deeper, recurring issues from just one bug and underlying problems.
Could this be Fable 5.5? My bet is next week, assuming the launch stays on track.
1
1
24
1,723