Model: Any NAI based one. If it takes tags it will work.Tag to use: greyscaleNegative prompt: defualt NAI will work beautifully. But not having anything somewhat works alright. Dataset size: 2,486 (Everything I could get with some cleanups on my part)So lads and lassies this is my first LoRA. I'm learning as I go but this one came out nicely. This was trained on anythingv3 (V1) or NAI (V2).And I should say this LoRA is fully capable of creating nsfw of all sizes (if prompted) if you know what I mean. So have fun.

Nanashi (NLO) 

Soon... I said oh well. Anyways forgot the amount of repeats for this one. But this retraining came out better. Should Look closer to the dataset.Trained on 768 resolution. Epoch: 12 V1 was a bit too over cooked.

Files with SHA256: b6dac3d9e411c83d6f33850bf864fdcee218740b33de407f370f267e4f4b9de4