The models don’t change.

tornikeo 2 hours ago | parent | next [-]

On paper. There's huge financial incentive to quantize the crap out of a good model to save cash after you've hooked in subscriptions.

▲

armchairhacker an hour ago | parent [-]

And there’s an incentive to publish evidence of this to discourage it, do you have any?

	▲	TeMPOraL 14 minutes ago \| parent [-]
		Models aren't just big bags of floats you imagine them to be. Those bags are there, but there's a whole layer of runtimes, caches, timers, load balancers, classifiers/sanitizers, etc. around them, all of which have tunable parameters that affect the user-perceptible output.

▲

esskay 2 hours ago | parent | prev | next [-]

Real world usage suggests otherwise. It's been a known trend for a while. Anthropic even confirmed as such ~6 months ago but said it was a "bug" - one that somehow just keeps happening 4-6 months after a model is released.

▲

fer an hour ago | parent | prev | next [-]

They do. I'm currently seeing a degradation on Opus 4.6 on tasks it could do without trouble a few months back. Obvious I'm a sample of n=1, but I'm also convinced a new model is around the corner and they preemptively nerf their current model so people notice the "improvement".

▲

stavros an hour ago | parent [-]

Make that 2, I told my friends yesterday "Opus got dumb, new model must be coming".

	▲	arcanemachiner 19 minutes ago \| parent [-]
		I swear that difference sessions will route to different quants. Sometimes it's good, sometimes not.

▲

girvo an hour ago | parent | prev | next [-]

I think the conspiracy theories are silly, but equally I think pretending these black boxes are completely stable once they're released is incorrect as well.

▲

pixel_popping 2 hours ago | parent | prev [-]

Oh yes, they do.