Post #3942092
2026-07-19 18:03 UTC
I'm curious if this is a controversial opinion:
Claude Opus ~ 4.6 is good enough for just about all coding problems.
4.5 is ok, but I do like 4.6 better. 4.8 is even better (and what I mostly use), but not a lot. If I were forced back to 4.6, it'd be fine. Fable is… slow. I really don't get why people care about Fable.
My point: Maybe we're kind of done? And if we don't need to train a new model 3 times a year, and just need inference, maybe we don't need a gazillion data centers?
But… 1/
Replies (1)
-
@nuthatch@infosec.exchange 2026-07-19 20:56
@cocoaphony@mastodon.social anecdotally, I never used the term “one shot” until I ran Fable through some personal projects. It seems to get many details “good enough” without a lot of oversight. It is slow. Fable seems to better infer under-specified requirements and makes good guesses the first time. Opus is effective but will sometimes break stuff that was previously working. It gets “distracted” if you don’t manage context? I’ve seen people suggest planning with Fable and executing with Opus. Hope to try that approach some day on bigger problems. I do try to use the simplest/cheapest/most focused model available. I’m writing one-off Swift prototypes, not solving hugely complicated problems.