Build a palace
Pick a station, or read a photo of your room for personal loci.
Read my room
Upload a photo; the vision model reads 4–6 loci from it to personalize your words. You'll still pick a station above for the visual room — this only changes what the words are.