 | Meta released Muse Image this week so I ran it against OpenAI's gpt-image-2 and Google's Nano Banana 2. I used the same source duck image and the same edit instructions prompt for every model (unchanged → blue → face away → glass → wireframe → hat-on-ball → "FRENZY" text → standing on a mirror with a correct reflection). The transformations go from easy on the left and gradually get harder. I ran 3 runs per model. Each model was then scored using a fixed 27-point rubric. One of these rows is Meta's new model. The reveal and full scores are in the comments. [link] [comments] |
Predictive technology: computational social science
![[link]](https://i.redd.it/aw7fuhhwq2ch1.png){kind=link}