MOSS-Image AI turns one line of text into a finished poster, menu, label, flyer or comic, with English and Chinese words spelled as you typed them and placed where you want them. Put the exact words in quotes, or turn on layout mode and drag boxes for the headline, price or logo. If one word, date or price is wrong, you can select that part of the image and change only that text instead of remaking the whole design. It runs in the browser with no install. The free trial is a one-time grant of 6 credits (2 without sign-up, plus 4 after email verification or Google sign-up). After that, one-time credit packs start at $6.99, with no subscription. MOSS-Image AI is an independent site, not affiliated with OpenMOSS.
The problem I set out to solve is a more user-friendly way to edit text in a specific part of an image. Most image generators are fine at pictures and bad at words: a headline comes out misspelled, or a date and price land in the wrong place, and the only fix is to remake the whole image. I wanted to select the words that are wrong, type the replacement, and leave the rest of the design alone.