For web app generation, gpt-5 seems to route my prompt to the reasoning model. Interestingly, gpt-4.1 often produced more creative or aesthetically interesting designs. In some cases, a bit of controlled hallucination (via temperature) is actually desirable, especially when generating UI ideas. Anyone else seeing this tradeoff?
All of these models generally produce hideous sites that because for some unknown reason we all decided shadcn and tailwind-with-no-design-system should define AI generated UI.
I use image generation for UI layout and have Claude implement it with an actual UI library: usually MUI with theming, but honestly anything with a sane grid system is better than loose tailwind.
I hear you, though I’m less focused on the choice of UI libraries and more on how little control we have over how prompts are routed internally.
Sometimes I want a model that reasons deeply, other times I want one that is more creative. Right now, it feels like gpt-5 forces everything through the same pipeline, even when a different mode would be better suited to the task.
