When you ask Claude, it says Claude is the best, and when you ask GPT, it says our Codex is the best! ㅎㅎㅎ
Copilot's token usage is absurdly high, and despite it being seamlessly integrated into the IDE, I'm trying to overlook it.
Most AI assistants' responses suggest that if you spend about $100 a month, you can use them comfortably without affecting your work. ㅋ
(But when I actually tried it, Copilot wasn't the case. I'm not sure how other tools would be if they had limits too.)
Anyway, all these responses feel like old-fashioned street medicine peddlers. I keep getting the impression that if you take this medicine once, all ailments will be cured.
I like GPT's newly released Thera because it performs well, but Claude's code quality seems much better, so I'm still torn.
