While working on a project, I kept jumping between models like GPT-4 and newer ones like Claude Sonnet 3.5, always thinking, "Maybe the other model has a better answer." I wanted a way to compare responses side by side.
Success story sharing