This LoRA creates an afterimage effect, making parts look like they are rapidly moving back and forth between two positions.
In v0.0.1, you can expect the effect to work reasonably well with basic positions, but it is currently a bit finicky and highly experimental. Due to gaps in the training data, the effect rarely works for fellatio or handjobs (these are included in the dataset, but with so few patterns that they are barely reproducible). It also completely fails to activate from a side angle (from side).
I plan to address and fix these issues in the next version.
Description
tongue_v0.0.0 (Initial) Trained on anima_baseV10. This version uses 10 cropped, close-up illustrations focusing specifically on the tongue afterimage area as the dataset.
Important Notes:
Composition Bias: Since the dataset consists entirely of cropped close-ups, the model tends to lean heavily toward close-up compositions if you use too few tags.
Model Compatibility: The afterimage effect may barely appear on models other than
anima-base-v1. In such cases, you can try forcing the model to "remember" the training by mixing in prompts like(multiple tongues:0.8),afterimage, andmotion blur.
dim16/alpha8, unet 1e-05(lion8bit/cosine_with_restarts[2]), 10images / 1400 steps