The total dataset is made of 100 pictures, and the training has been done on runawayml 1.5 and the new VAE, with 12k steps (poly LR1e-6)
The recommended sampling is k_Euler_a or DPM++ 2M Karras on 20 steps, CFGS 7 .
The total dataset is made of 100 pictures, and the training has been done on runawayml 1.5 and the new VAE, with 12k steps (poly LR1e-6)
The recommended sampling is k_Euler_a or DPM++ 2M Karras on 20 steps, CFGS 7 .