Bandit retweeted
GPT-6.2 is getting roasted on X
meanwhile, Astra and Sol just took #1 and #2 on my object detection benchmark
so I have some awkwardly positive results to share
↓ benchmarks and examples
I think this is the first company-level acknowledgement that the UX for GPT models has become fairly poor. That's a good sign, but unfortunate that we won't see improvements in the near term given the current pace of releases.
Replying to @sama
Although the new models are very smart, we have more work ahead of us on making the models nice to work with, and will prioritize that for upcoming releases.
But even if it takes a little extra work, the intelligence and capability are extraordinary.
39
okay so wen Bel
OpenAI simply HAVE to understand the shortcomings of their models, right? Are they doubling down on this direction, hoping RSI will let them fix it later?
Replying to @emollick
This is the first time since GPT 4.5 that a frontier release from OpenAi is not a clear step forward. I’m willing to bet that 6.2 is amazing at some types of complex work, but in my experience & that of others I know, it was often frustrating to use.
1
4
649
The board did the right thing to remove him. He reportedly lied to the board and said there was only one affair, then the dam broke.
The board can, at its discretion, also hire him back on probationary terms.
Feels like the latter is growing in possibility.
Over the past week, I’ve been overwhelmed by the love and support of friends, industry colleagues, and people I've never met. Thank you to everyone who has shown kindness to my family and me. It means the world to me.
I came to the United States 53 years ago, and becoming an American citizen is one of the greatest gifts and proudest moments of my life. NVIDIA has been my life’s work. Building it into the company it is today alongside my colleagues has been a privilege and an honor. Nowhere but in America could a company like NVIDIA be possible. I love our country, and I love NVIDIA.
I did not live up to the standards expected of everyone at NVIDIA, and I have made serious mistakes in my personal life. I created problems for the Board that I deeply regret. I own my mistakes completely, and I am committed to cooperating with the Board’s independent investigation.
Since my departure, I’ve seen an outpouring of calls from NVIDIA’s employees, investors, business partners, and customers for me to return as CEO. Many continue to reach out to me directly as the company navigates complex technology, market, competitive, and geopolitical dynamics. NVIDIA is a once-in-a-generation company. If the Board of Directors decides that my return is in the best interest of our shareholders, our employees, the company, and America, I would be honored to answer these calls and return as CEO.
Thank you again to everyone who has shared their good wishes. I am profoundly humbled by your support. To my family: I love you, and I will spend the rest of my life being worthy of yours.
Jensen
98
Assuming this must be completed prior to release, whatever model this is may not debut in public until December or January.
1
36
Speaking as a technical buyer for these kind of services, LaJ benchmarks are inherently weak as a signal for whether the product is suitable for industrial-grade processing.
I need concrete, repeatable benchmarks so I can make an informed decision. Rules-based is needed.
This benchmark came out a few hours ago. We were able to audit this quickly because modern frontier LLMs are great at comparing parsers.
There is no reason to be using rule-based benches anymore, LLM judge is the way.
parsejudge.com/parsebench-2
1
39
Okay now the FYP is serving this up and just outright trying to bait me.
The dataset is public! As a potential customer I can inspect and verify it myself! LaJ has a useful place but this attitude towards an open benchmark shows a very poor understanding of enterprise needs.
look, i get it. benchmarking is inherently a hard task. you take human expectations, which are qualitative, and then bring them into quantitative land.
any metric you choose for quantitative measures on a benchmark is going to be inherently motivated and subjective. that's not a bad thing when you, the person using the tool, is setting the standard. it's a TERRIBLE and unfair thing when the tool makers set an arbitrary one.
that's why parsejudge.com exists. so that you can set your own benchmarking standard, on your own documents, and then immediately see the quantitative way it matches.
take a look at parsejudge.com/parsebench-2 for a good example of where traditional benchmarking falls short.
1
31
It's simply because it communicates better, so for a wide range of work it is more effective than Sol/Astra since the users can more easily understand it and maintain tactical alignment.
Now, if OAI models improved on its own user understanding, we'd have GPT-Crack on our hands.
Replying to @JamesMalsawm
Pretty wild to me that GPT 6.1 Terra gets compared to Opus 5.5. No doubt that it's a great model, but it's not meant to compete in the big leagues. We've got other models for that.
47
Can't wait for three years from now when everyone is riled up about the usage hour limits for the bi-pedal OpenAI Dot Bots being cut in half by cyborg Tibo.
46
I love Codex and the Codex team, but man what's going on? It was clear from the jump that the 28 days is partially a play for time, maybe more compute coming online or a model finishing training, but you get the sense something very challenging is going on behind the scenes.
1
38
if enough of us pull out the GPUs from our gaming desktops and offer them personally to Tibo, we might save our usage
I'll bring my RTX 3060. that's gotta be good for like, a couple Luna turns right?
46
It's unfortunate because Terra would appears to close an actual gap in OpenAI's lineup that was badly needed. Forcing business users to choose between Sol and Luna meant Sol would be chosen, which was overkill when a Terra would be fine.
Now, hopefully the release redistributes model usage and relaxes the compute constraints. If another part of Abilene is coming online this quarter, then one should expect OpenAI to be able to finish the year stronger.
OpenAI released a model that is intelligent, accurate, and fast. So, why is everybody upset?
Well, it's pretty simple: everyone in the AI community is currently expecting the drop of a heavily intelligent frontier model like Fable 5.5. We also know from OpenAI researchers that GPT-6 has been obsolete for months, plus OpenAI has been teasing math breakthroughs with their new frontier model. Everyone in the current AI diaspora is expecting a frontier model.
So when you release Terra, the dichotomy between what people are expecting and what they are getting is quite obvious. Even if the release itself is actually really good, innovative, and very efficient, people like me and others on X are not hungry for an efficient, intelligent, small model release. Also, the numbers are pretty close to Sonnet 5.5 in terms of cost per task.
It's pretty obvious to me why the sentiment is against OpenAI right now. It feels like we're being cheated out of frontier models, and the gap between what the labs have and what the public gets is unfortunately growing (which I know to be true because it is).
33



















