aidoll.fun shop was successfully tested by first two buyers! Everything works well, except email notifications. So please don’t be disappointed if you buy something and don’t recieve any email.

Is AI artist, creator of cute stuff and provider of extra networks for stablediffusion.

Description. Embedding trained for 2000 steps on ‘v1-5-pruned-emaonly’ checkpoint, so it should works well with most of v1.5 forks and mixes.

Is AI artist, creator of cute stuff and provider of extra networks for stablediffusion.

Is AI artist, creator of cute stuff and provider of extra networks for stablediffusion.

Is AI artist, creator of cute stuff and provider of extra networks for stablediffusion.

Description. This LoRA trained using kohya_ss on 60 images 768×768 up to 1500 steps with handmade captions and without regularization images. Source model: SD15, learning rate: 0.00001, Precision: fp16, LR Scheduler: constant_with_warmup, LR warmup 30%.

Description. This LoRA trained using kohya_ss on 60 images 1024×1024 up to 7200 steps without captions and with regularization images. Source model: ‘Source model: SD-XL_1.0-base,’ so this LoRA should works well with most of SDXL family checkpoints.

Description. This LoRA trained using kohya_ss on 60 images 1024×1024 up to 6360 steps without captions and with regularisation images. Source model: ‘Source model: PonyDiffusionV6XL,’ so this LoRA should works well with most of Pony family checkpoints.

Description. This LoRA trained using kohya_ss on 60 images 1024×1024 up to 12000 steps (50 epochs, 240 steps each) without captions and without regularization images.