cududa

Please for the love of god don't tell me the Codex sandbox is their actual eval harness sandbox?????

I maintain my own fork of Codex for "fun". Whenever I look at the sandboxing churn they're doing every release, as someone who used to work at Microsoft on Windows, my reaction is usually: https://c.tenor.com/vTzzhTiypwQAAAAC/tenor.gif

This is really cool! I’m becoming convinced the optimal UI to engage with agents, long term is going to be something spatial. No idea shape that even takes, though I really feel what you’ve made might be Xerox PARC days in terms of metaphor maturity, but there’s some real new seeds of “obvious in retrospect” ideas here. Thanks for conceiving of and building this!

Right? I also quit Claude Code and switch to Codex over that. Now I’m trying to figure out how I could make an extra $65,000 to never have to be concerned about this nonsense again. I know the economics of using open router etc…

But I’m reminded of ~2008 and the rise of “the cloud” as a marketing term that seemed to me to be a cover for dropping an expectation of rich clients, increasing a companies margins around subscriptions that would chip away at local ownership.

Then I got offput by the zealotry and absolutism around “true FoSS”, told myself I was young and moved on.

And really, a lot of subscription models I kind of can appreciate/ tolerate. Might be irksome but whatever, I get that software is expensive to make and it’s not fair in 2026 to value a yearly upgrade of Photoshop at $200. The capricious UI changes to things that’ve worked for 20 years and they take away say the classic color swatches altogether - silly and dumb.

I can use another professionally necessary tool I pay $200/ mo for, Codex, to whip up a classic swatch plugin.

Is that $200 a fair price for my token usage? I think an extremely heavy month I might’ve used a billion tokens?

But that right there is the problem. They have no idea what, specifically, profitability looks like and are going to be pulling endless levers for … I genuinely have no idea how long - at least through 2030/2032 if we tea leaves their debt obligations?

I don’t want to think about any of that. At all. I don’t want to spend time evaluating model preference and degradation and updating the nuances of how I “speak” to an AI because there’s some mystery backend experiment running on the output I use to produce functional outputs — ie the actual products I get paid to build/ maintain.

AI’s something between a tool and coworking companion, and the capricious “personality” changes due to playing with poorly understood and knobs and levers at the inference level - is maddening. To that end, I want a box in the corner I can point to and know exactly the quality of outputs that no one but myself modifies.

Embedded devices absolutely need DDR3 and DDR4

I mean it’d take minutes of research to realize people are successfully and efficiently running 4-bit quantized GLM 5.2 on MacStudio 512GB M3 Ultras at over 60 tok/s. K2 2.7 is quite literally designed for 4 bit quantization and runs even better.

This is already a thing

Ah yes but you can shunt the costs of that off into a public works/ taxpayer funded infrastructure project

How do you figure that? “also a reminder that as soon as Chinese models take the lead, they will switch to closed source too”

What specifically about their release strategy “reminded” you of that conjecture?

The premise that they only open source the models … because it somehow helps them leapfrog American labs, and once they actually can leapfrog them, they’d close source them, doesn’t really track for me. Am I missing something?

I mean I think we need our own domestic open weight labs. I just don’t particularly understand the point you’re making

I’ve been exceptionally displeased with Claude Code since end of February and switched completely to Codex in April. The blasé way in which one person (Borris) capriciously changes the system prompt multiple times a day, also no longer writing his own prompts (whatever that means).

That, the 5 different secret levers you have to pull to make it not stupid, the fact you hs e to go to the guy’s twitter account to find all the un-dumbing features and flags that aren’t documented anywhere else. That they decrease thinking budgets silently when they run out of compute instead of announcing the rationing, and gaslighting users at every step of discovery. The fact that internally they have their own coding harness and don’t use Claude Code primarily. The lack of formal evals and consideration for millions of users collective hundreds of millions of hours of investment in their workflows — that’s all off the top of my head, let me tell you how I really feel about what they did to Claude Code..

I adore gpt5.5 and maintain my own codex fork - but I have no idea how long I’ll get this performance / cost - I know it won’t be forever. I’d like to know precisely how much it’ll cost in hardware to run a gpt5.5 open source model locally. Hell a lifetime license to a model I can run locally is also be open to.

But I like building my own tools, from software to physical shop tools. I like being able to rely on my tools.

More responding here to the assertion that this is blowing up due to Fable.

It doesn’t matter if people have to suddenly live by gas turbines that run 24/7 because why again? Can you repeat that last part back to me but say it a little dumber for me?

This might be the best color palette generator I’ve ever seen. I used to work in Operating Systems, and trying to get a good color palette from a photo is HARD. A lot of very smart very well paid people have dedicated years of their life to this type of thing. Really fantastic work.

If the author of the blog post ever comes across this thread/ comment, bravo and I hope you feel pride in your work and I’d go so far to say discovery.

I see a deep tree and a shallow tree in the two screenshots, representative of entirely different approaches.

You’ve never used the API version versus the $200 plan and set the two at the exact same task, have you?

Huh, seriously? Have you ever worked in an office? Perhaps your mental picture of what op is describing might be misaligned? I just always assumed it was a rarer/ more disciplined style some people had

“Well, did it work for those people?”

“No, it never does. I mean, these people somehow delude themselves into thinking it might, but……

…But it might work for us!”

Remember when they shipped that version that didn't actually start/ run? At work we were goofing on them a bit, until I said "Wait how did their tests even run on that?" And we realized whatever their CI/CD process is, it wasn't at the time running on the actual release binary... I can imagine their variation on how most engineers think about CI/CD probably is indicative of some other patterns (or lack of traditional patterns)

As someone that used to work on Windows, I kind of had a vision of a similar in scope e2e testing harness, similar to Windows Vista/ 7 (knowing about bugs/ issues doesn't mean you can necessarily fix them ... hence Vista then 7) - and that Anthropic must provide some Enterprise guarantee backed by this testing matrix I imagined must exist - long way of saying, I think they might just YOLO regressions by constantly updating their testing/ acceptance criteria.

Why not provide pinable versions or something? This episode and wasted 2 months of suboptimal productivity hits on the absurdity of constantly changing the user/ system prompt and doing so much of the R&D and feature development at two brittle prompts with unclear interplay. And so until there’s like a compostable system/user prompt framework they reliably develop tests against, I personally would prefer pegged selectable versions. But each version probably has like known critical bugs they’re dancing around so there is no version they’d feel comfortable making a pegged stable release..

That guy has his own form of AI psychosis

When you say it’s not a massive codebase, I’m curious, what are you comparing it to?

Good. It’s always insane to me that they get 1% of the iPhone CPU cost of ~$68 or something there around.

There was a lawsuit in 2020 or 2021 where some evidence was unsealed showing ARM gets 1% of the CPU cost. I can’t recall how that CPU cost was calculated - but I believe that was a part of their deal through the early 30’s. That’s less than a dollar per iPhone.

Sure but that doesn't mean it'll perfectly retrieve information it's trained on. There's a lot of conflicting sources, hallucinations, etc.

Fantasy price for your personal usage or the personal usage of most average consumers/ software engineers, sure. It bears repeating: you're not the target audience.

They've been in the display game a long time. For people that need the product capabilities for their specific job, like color grading, they seem to price them quite well, given everywhere I used to see $30,000-$50,000 reference monitors, I see Studio Displays now.

Thanks for sharing. Very … interesting. Just trying to understand why the heck would React be the best tool here?

I just get all excited whenever anyone brings these books up, remembering the first time I read them.

Correct. If you try to create a coding agent using the raw Codex or Claude code API and you build your own “write tool”, and don’t give the model their “native patch tool”, 70%+ of the time it’s write/ patch fails because it tries to do the operation using the write/ patch tool it was trained on.

I came across this the other day. If you go to the photos tab on the @grom account, it’s an endless stream of women in bikinis or in explicit poses.

Apparently all anyone has to do is reply to any photo anyone posted of themselves, is something like “@grok put her in a bikini” — or more obscene demands, and it just spits out the image. It’s kind of disturbing how simple it is

Google or any open source map product. And actually, if we use the SCOTUS approved DOJ v MSFT consent decree as precedent, any app that can't use this private API component would be an impacted party.

I'm an antitrust nerd - 20+ years since I made my first PACER account as a teenager to get documents from interesting cases..

95% of what people call "anticompetitive" or "monopolistic" has no legal bearing. People don't know the legal definition of those words and bandy them about based on vibes.

This however, is a very very clear case of violations of precedent. If we look at Microsoft's final judgement https://www.justice.gov/atr/case-document/final-judgment-133 see F(1)(a), H(2)(b), while these stipulations haven't been applied to Apple, if I were in a market dominant position, I'd be super careful about capricious restrictions like the example undocumented API, and behavior that mimics patterns of activity that were seen as actionably sanctionable to similar market dominant forces

I wasn’t previously aware of Ukrainians flying drones directly into helicopter’s rotors, but that does make sense.

I obviously know very little about battlefield armaments so forgive the potentially stupid question - but have drones made combat helicopters obsolete?

Which part of the CHIPS act says companies receiving funds have to give the government 10% of the company to continue receiving funds?

Where in the bill passed by Congress does it say taking funds entitles the government to 10% of the company 3 years after the fact?

Did ChatGPT also disable his vehicle?

Great metaphor