I tried that for the GPT-5 launch - a self-improving loop that renders the SVG, looks at it and tries again - and the results were surprisingly disappointing.
replyI should try it again with the more recent models.
I see, thanks. I guess most current models are not yet trained for this loop.
replyCould you please try with Opus 4.7? I think there's a chance of it doing well, considering the design/vision focus.