I‘m not sure that’s true with the latest models. o3-mini is good at analytical tasks and coding, and it really sucks at prose. Sonnet 3.7 is good at thinking bu

I‘m not sure that’s true with the latest models. o3-mini is good at analytical tasks and coding, and it really sucks at prose. Sonnet 3.7 is good at thinking but lost some ability in creating diffs.