Using both daily now. Claude feels better at long refactors and reading existing code. GPT-4o feels faster at short fixes. Curious what others think — especially if you've compared on a specific task.
Sign in to reply to this thread