But sounds like Opus and GPT are interchangeble for your case.
Both have good models. My workflow these days is usually Opus plans and codes, then Opus and Sol review, finally Opus verifies the review and applies fixes.
Both have good models. My workflow these days is usually Opus plans and codes, then Opus and Sol review, finally Opus verifies the review and applies fixes.