Part of my facial expressions series. All of these LoRAs are designed to mesh well with each other and have low impact on your image framing.
Good clean fun. Thanks to @TheAiHelper for the idea.
You'll get slightly different results on winking (more eye focused) versus winking expression (more full-face focused with a wry smile). You may need to prompt way harder than some of my others. The effect holds up pretty well but it needs a little cajoling sometimes, it seems a little more seed dependent than some models, if it draws it's great, but if not it's really hard to bring it out.
Seems to work especially well with animated models, but only when the full winking expression is used.
Starter prompt:
<subject>, winking expression
<subject>, BREAK (winking expression:1.2)