Diffusion-Series

This is my Diffusion-series, including all my understanding of diffusion models

My notes on diffusion models, from the basics to generating in a single step.

DDIM: making the DDPM sampler non-Markovian and deterministic, so it can skip steps.
Flow Matching: diffusion as a velocity field along a chosen path; why straight training paths still give curved sampling paths.
One Step: consistency models, flow maps, MeanFlow, and distilling a many-step model into a one-step generator.

Jan 2026 Diffusion-Series Diffusion-Series 3 - One Step: Consistency Models, Flow Maps, MeanFlow, and Distillation How to generate in a single network evaluation: the flow map that every 1-NFE method learns, the from-scratch line from consistency models to MeanFlow and drifting, the three families of distillation, and where the ImageNet numbers stand. 17 min read
Jan 2026 Diffusion-Series Diffusion-Series 2 - Flow Matching: Diffusion as a Velocity Field From DDIM's deterministic sampler to flow matching and rectified flow: what the network learns, why the loss works, how it relates to epsilon-prediction, and why straight training paths still give curved sampling paths. 8 min read
Jun 2025 Diffusion-Series Diffusion-Series 1 - DDIM: Making Diffusion not Diffusioning This is a simple explanation of DDIM, a fast sampling method for diffusion models. It is based on the original DDPM, but it is not a diffusion model anymore. 4 min read