14,235 karma · joined May 6, 2012
Normally I run the models that the harness is intended to support. You can't run Claude models anywhere else besides Anthropic's own CLI and desktop harnesses -- or at least you're not supposed to -- without paying out the nose for API tokens.
I do use DeepSeek 4.1 Flash 1017 in Claude Code when I want an abliterated model for security-related work or unhobbled browser control, though. I know it's possible to get OpenAI models working in CC, but I haven't bothered. Gemini I'm not sure about.
Codex is pretty close in a lot of ways, as the other poster says, but Astra consumes tokens at a higher rate, and when it compacts it completely lobotomizes itself. That used to be a problem in Claude Code as well, but it has been largely solved (although at the cost of making the compaction take much longer.)
Antigravity reminds me of the old comic strip where Firefox and Chrome are duking it out, while Internet Explorer is sitting in the corner wearing a helmet and eating out of a jar of paste.
Yeah, it pretty much does. CUDA is why they are positioned where they are. Other players like AMD and Intel could have done that, but they didn't, and if they had, they wouldn't have persevered for 10+ years with essentially nothing to show for it.
Whether CUDA will continue to maintain dominance is less clear. It may become a literal victim of its own success, as CUDA-based models gain the ability to migrate themselves to new platforms.
Fable is great, certainly, but it's no better at writing or reviewing code than the top-end Carbon or Astra models. To the extent Anthropic has a moat, it's Claude Code. Obviously that's not much of a moat, but their competitors seem content to camp on the other side of the drawbridge for some reason.
Ferrites do nothing for low-frequency impulse noise, which is what this type of issue usually comes down to. They are there to pass FCC/CE tests, and that's it. Optically-isolated cables can indeed be a good solution, but not many of them support USB 2.0 HS unfortunately (which my stuff needs). Might be the right answer here.
("No background," LOL. To answer your question, Demis Hassabis won the goddamned Nobel for his work on AI-based medical research, specifically protein folding. Is that enough "background" for you?)
But yes, this is stupid. No one who is inclined to build a biological WMD is going to be more than mildly inconvenienced by supply restrictions. DNA isn't plutonium.
What have you done to cure cancer or prove the Riemann conjecture, anyway? Oh, wait, you answered that one already. Guess we're done here.
See Dan Gelbart's recent video on that very subject: https://youtu.be/v5d8nwgiSYI?si=iM9AFZiRTjxu4OuZ
Probably unhealthy, but dammit, it's just too much fun.
Let's drill down on this point, since it's the one you chose to preface your argument with. Who, exactly, has been "claiming AGI," whether "for ages" or otherwise? Cite your sources: has Anthropic in particular claimed AGI?
The designer could just as easily have closed the same feedback loop with Fable or Astra.
Perhaps other sites beckon.
What have you done to cure cancer or prove the Riemann conjecture, anyway?
Really?
There have also been many incredible breakthroughs in math in recent years without AI. But things are different now.
If the AI money had gone into that, I suspect we'd have better results.
Otherwise known as "eating your seed corn."
That's about all you can say for sure, especially about a topic that is so broad as to be virtually meaningless.
The state of the art doesn't matter. Only the first couple of time derivatives matter.
I don't think that case can be made. I'm a "good coder" and the best LLMs are now better at it than I am, in essentially every respect that matters. I can't think of a test that I could reliably pass that Fable or Astra would reliably fail.
I don't know who they think they're fooling.
Edit: Fable 5.2, whatever. There you go.
The real problem isn't going to be the job losses themselves, it's where we'll get the next generation of skilled workers. There will always be a vacuum at the top, regardless of how good AI gets.