4 ms·
IME I can't trust it to write it's own plans from a spec, but if I give it a detailed execution plan written by Opus, it's fast and cheap (if chatty) in executi
by coredog64 2mo ago
IME I can't trust it to write it's own plans from a spec, but if I give it a detailed execution plan written by Opus, it's fast and cheap (if chatty) in executing it.
- stavros 2mo agoThis is what I do, and it works fantastically well. Just make sure you have Opus/GPT review after.
- polski-g 2mo agoInteresting. I use Flash for making the plans and GPT for execution.
- Kadin 2mo agoDepending on the language you're writing in and the problem domain, the smaller models can do dramatically better or worse. I suspect in the future we'll see language-specific small models. "Coding" is still pretty broad as an activity. It'd be nice to be able to load up a model specific to, say, class-based Python and run it on-device.
- eru 2mo agoHarmonic's Aristotle is a sort-of language specific model for Lean, if you want to see the future you described today.
- jatora 2mo agoflash for plans?! i don't understand why you wouldnt use something far stronger for the most load bearing point of the project
- tokai 2mo agoPrice obviously
- ricardobeat 2mo agoThere aren’t many “far stronger” models than Flash 0731 now, it’s only beaten by Claude and OpenAI models at high/max effort, and everything that matches it costs 5x-10x more.
- arcanemachiner 2mo agoYou're doing it backwards.