kristjansson

It's like using bathtubs as a unit of measure for large volumes of water. Yes it communicates the scale, but abjectly fails to communicate the relative in-cohort comparison i.e. is this a lot of power for an industrial user? Is it wasting power /generating more emissions relative to other similar applications or not?

This is the right perspective. It costs as much as one additional tool call in the conversation. It’s a bit weird to do it on a timer, which inflates a bit over “just a tool call”, but not by much?

yes yes as everyone knows there are exactly 112,530,874 jobs in the US, and will never be any more or less, so the labor market is exactly zero-sum

> for free because they think it is good for humanity

I like Qwen as much as the next person, but that's a bit to credulous. Those models would not get the compute allocation nor distribution they do if it was not aligned with broader policy.

There's a great bit of shortform video content floating around that shows a stammering younger guy in a welding mask:

  > Do you have two pieces of metal that need to become one? 
  > But if it became  two pieces again it wouldn't be a big safety concern?
  > If that's you call Bob's OK welding!

That's Jev, and ~all zero-shot classification-by-LLM. It's universally OK. And that's great, for the tranche of problems where OK is good enough. But anywhere OK is not good enough, one would be advised to use or build something somewhat more robust.

> Installing the update isn't enough. A running process keeps the old decoder until it restarts, and a container keeps it until you rebuild the image on a patched base and redeploy

Claudeslop aside, it really is a whole new world we’re in now, where the audience of a run-of-the-mill RCE CVE announcement really does need this stated.

> rounding error

Not really? That’d be ~1% of global revenue and ~2% of opex? Sure they could pay it, but it’s not a negligible amount.

I need to see SQLite implemented in PowerShell

No-stakes work is ... worthless? If no one cares about the work enough to have any input or requirements for the results ... sure it might be unrecognized genius that's going to take the world by storm but the absence of interest is not a good sign!

The CLI has been pretty good on this score, way better than the App. Has been…

not to mention the daemon mode breaks a bunch of assumptions one might have about about codex CLI e.g. that codex CLI inherits the env vars of the shell that invokes it.

—no-daemon is the way, but still feels like the noose is tightening …

Delaware law is not so much favorable as it is well-trodden. Of all places here we should understand the value of the network effect.

> The key mistake people made was thinking in terms of their own workflows and own local stacks

We need a dunning-kruger for empathy: people least able to imagine situations not their own trying hardest to influence others.

The user cares about the distribution of outputs. That distribution is structurally determined by the distribution of inputs (i.e. prompts), the weights of the model, and (these days) the dynamics of the harness guiding successive generations.

The only way to characterize whether a choice is 'right' is to characterize the output distribution (i.e. evals)! Changing the underlying weights necessarily invalidates whatever characterization may have been done. One may assert that one's harness regularizes outputs back toward the desirable distribution, or one may hope the different weights induce a sufficiently similar output distribution.

But no, one should not be completely agnostic to the choice of weights just because there's some nondeterminism.

Well, time to set down the glass beads and dive into a an alpine lake.

With MTP and FP4 I max out at 30ish t/s on mine. Without MTP or in regimes where the drafter performs poorly it’s about 10 t/s. FP8 is about half that

They made the coding plan a bit better toward the end, but it was pretty tough to use throughout.

Seems like an Amdahl’s law of inference economics? there’s so much compute relative to SRAM on the chip and shoreline bandwidth onto the chip that caching buys ~nothing? The contended resource is SRAM and a given token of context needs just as much as another.

> one TLD does not operate at the most fundamental level like the others;

contrary, this is one TLD _ceasing_ its idiosyncratic mode of operation.

Of course there's some trim that has to go into the grinder, but we could do a much better, more respectful job of using the animal as it was instead of homogenized lumps of 80/20 or 90/10. More beef cheek in the butcher case and homemade cabeza, more lengua, more shank and round and plate, more heart and tripe and liver, etc etc etc. Buy a subprimal and grind your burgers at home from a single animal.

This is more a culinary and ethical argument than efficiency or safety. One should use the whole animal, in all its varieties, because they're delicious. Grinding otherwise edible parts of an animal killed on one's behalf because one disfavors (or is repulsed by) them seems to me much worse than the killing. Eat - and enjoy! - all of it or don't kill it.

I don't know if you're a bit out of the loop, but ... meat is very much not at issue in this case.

> grinder

There's your problem. Ground products are completely dispensable.

> LLMs could

This is going to end up like the Law of Headlines, isn't it? "Do x, y, z Cure All That Ails You?" ... no but we got you to read the article. "LLMs _could_ x, y, z" ... but they don't because they're programs, not magic.

It seems impossible for this to be a true statement. and yet.

What’s next? $20m series B in ByBy, a horizontally integrated trash disposal, concrete shoe manufacture, lye bath, and pig farm operation?

The operative part of schoolwork is … not the work part? Of course school work is bullshit, in the sense that the teacher already knows what 7x9 is or who fought at Waterloo or whatever. The point is not the work product, it’s that the student produced it and has been incrementally improved by the experience!

“Passive Income? Bro, I need MASSIVE INCOME” as the tagline of a sports gambling operation … I guess “your desperation is our opportunity, give us your money you illiterate innumerate peasant rube fuck you” didn’t fit on the billboards?

hf download maintains each downloaded version in a commit-hash-addressed snapshots dir, with symlinks to the underlying files content-addressed in a blobs dir. It’s not quite git, but it is a stable reference path.

Just quantizing takes seconds-to-minutes, llama.cpp provides a nice tool[0]. Improving quality is then a matter of picking specific tensors to maintain at higher accuracy, checking on representative data, and repeating.

[0]: https://github.com/ggml-org/llama.cpp/blob/master/tools/quan...

I'm generally happy with the tests I ask it for, some of which are PBT. It's just the insistence upon memorializing every single change with a test. Maybe encouraging / forcing PBT will dissuade it?

And the tests. Oh god the tests. Personal recent favorite: I asked for some changes to a Dockerfile, which it did ably, and then promptly tested by writing a pytest module that traversed up to the root, read the Dockerfile, and checked that the added lines were present.