Trained on/generated with EveryDayArt v5. When I've tried using my loras on other models in the past, the result isn't very good. Maybe I should swap to a different base for training some day, but to recreate images like the ones in my posts you should definitely use that model.