Denoising loop / scheduler
Generation runs the same network many times, each pass removing a bit of noise under the scheduler’s plan.
Popularized inJun 2020Denoising Diffusion Probabilistic Models
A diffusion model is executed in a loop: starting from pure noise, the network predicts (and the scheduler removes) a slice of noise at each step, gradually revealing the image. The scheduler defines the noise timetable and update rule. Flow-matching variants learn a straighter path that needs far fewer steps. In the graph this whole loop appears as a single ‘denoise step’ stage that runs N times.
Adoption over time
Share of new models that have included a denoising loop over time.
See it in real models
Open any of these on hfviewer to find this block in the interactive architecture graph.
Comfy-Org/MiniMax-H3↓ 17.3M · ♡ 2kOpen in visualizer
MiniMaxAI/MiniMax-H3image-text-to-video · ↓ 4.0M · ♡ 4kOpen in visualizer
Comfy-Org/Krea-2↓ 2.2M · ♡ 465Open in visualizer
stabilityai/stable-diffusion-xl-base-1.0text-to-image · ↓ 1.6M · ♡ 8kOpen in visualizer
Lightricks/LTX-2.3image-to-video · ↓ 1.6M · ♡ 2kOpen in visualizer
stable-diffusion-v1-5/stable-diffusion-v1-5text-to-image · ↓ 1.4M · ♡ 1kOpen in visualizer
stabilityai/sdxl-turbotext-to-image · ↓ 1.1M · ♡ 3kOpen in visualizer
Tongyi-MAI/Z-Image-Turbotext-to-image · ↓ 794k · ♡ 5kOpen in visualizer
Lightricks/LTX-2.5image-to-video · ↓ 790k · ♡ 2kOpen in visualizer
nvidia/Cosmos3-Edge↓ 789k · ♡ 170Open in visualizer
black-forest-labs/FLUX.2-devimage-to-image · ↓ 655k · ♡ 2kOpen in visualizer
black-forest-labs/FLUX.1-devtext-to-image · ↓ 635k · ♡ 14kOpen in visualizer
Browse all 237 models with this in the catalog →
Related concepts
hfviewer renders the full architecture of 4,000+ Hugging Face models as interactive graphs. Hover any block to see what it does, with this model’s real numbers.
Browse all model graphs →