3 ms·
from https://arxiv.org/abs/2310.04378 https://arxiv.org/abs/2310.04378 it sounds like its a form of distillation of an SDL model. So im guessing it can't be dir
by quadrature 3y ago
from https://arxiv.org/abs/2310.04378 https://arxiv.org/abs/2310.04378 it sounds like its a form of distillation of an SDL model. So im guessing it can't be directly trained, but once you have a trained diffusion model you can distil a predictor which cuts out the iterative steps.
While it can do 1-step the output quality looks a ton better with additional steps.
- dvrp 3y agoYou can train it directly. This is from the paper “An LCM demands merely 32 A100 GPUs Hours training for 2-step inference [...]”
- T-A 3y agoLet's quote the whole sentence, shall we? An LCM demands merely 32 A100 GPUs Hours training for 2-step inference, as de- picted in Figure 1. [1] Now let's look at the caption under Figure 1: LCMs can be distilled from any pre-trained Stable Diffusion (SD) in only 4,000 training steps (∼32 A100 GPU Hours) for generating high quality 768×768 resolution images in 2∼4 steps or even one step, significantly accelerating text-to-image generation. The training mentioned in your quote is distillation: it requires a previously trained SD model. You could have told just by reading the introduction: We propose a one-stage guided distillation method to efficiently convert a pretrained guided diffusion model into a latent consistency model by solving an augmented PF-ODE. [1] Section 4.2 in https://arxiv.org/abs/2310.04378 https://arxiv.org/abs/2310.04378