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<s> TODO: get a larger more diverse RL dataset and filter it for artifacts</s> TODO: do larger ranks and 1536x training, as well as auxillary loss(es)

[Anima] Flash 8 steps cfg 1 distilled

epoch 80, now with 18 samples at 2.25MP in the dataset it's less style collapsed at higher res but idk if better too tired from testing right now i may upload a revised / SMA merged version try strength 1.0-1.25

Files with SHA256: 9511481ec62422b754483cd05ff600b266ea9db0dbbf41bd685bf81386aa447d