Ah, that reminds me of an article someone here linked to a few years ago: [1]. That claims the efficiency is -3.5 dB compared to a typical -1.25 dB dipole in dedicated wireless equipment. Not bad; the difference is only a factor of ~2 in power for x8 reduction in size.

[1] https://www.embedded-computing.com/articles/a-lesson-in-wire...