Although, this isn't so much a model as a dare experiment: I can train models on a single image. In the mobile game Backpack Brawl, a new hero was disabled, and I decided to first run a video of it through WAN 2.5, then cut them into frames and train the lore. The result is what you see on the screen.