> I can imagine a curve where for a given level of “intelligence” you either need model size, or inference time (“test time compute”), and can somewhat trade one for the other.
This is exactly what was shown on Luna/Terra/Sol tradeoffs - Luna requires much higher reasoning efforts to approach Terra/Sol on lower reasoning. Which is fine, of course, no complaints - but true.
This is exactly what was shown on Luna/Terra/Sol tradeoffs - Luna requires much higher reasoning efforts to approach Terra/Sol on lower reasoning. Which is fine, of course, no complaints - but true.