a distillation lora for anima, trained on ~100 images at 1024x generated by the model itself using 40 steps cfg = 4 er_sde from diverse captions
baked-in negative prompt: worst quality, low quality, score_1, score_2, score_3, blurry, jpeg artifacts, sepia
may or may not be underbaked/overbaked
works best with cfg = 1 and 8 steps for heun/2s samplers, 16 steps for 1s or multistep samplers, but you can use however many steps you'd like.
the training code is a modification of diffusion-pipe for RL/distillation + custom comfyui nodes for data generation
TODO: get a larger more diverse RL dataset and filter it for artifacts
TODO: do larger ranks and 1536x training, as well as auxillary loss(es)
Description
trained for 10 epochs on 93 images, 1e-4 lr adamw 0.01 WD, rank 2. this is a bit of a test lora but it performs really well at 16-30 steps and cfg = 1. trained at 1024 on top of base v10 but 1536 gens work too. has less style variance collapse than rdbt i think...