🌀 Diffusion – Maitreya

Forward and reverse processes, guidance, and the noise schedule - how a denoiser turns into a sampler you can control.

Fewer Steps: Follow the Curve, Straighten It, or Skip It

generative

diffusion

Four responses to the same fast-sampling problem, derived rather than listed. Including the result that surprises people: a mathematically better solver can be twice as bad as a crude one when you only have two network calls to spend.

Aug 1, 2026

Deep dive

What Acceleration Actually Costs

generative

diffusion

A paper reports 50 steps to 4 and calls it a 12× speedup. Your service may not see 12×. This post is about the gap: the things you pay for besides network calls, the accelerations that cancel each other out, and the failure modes that only appear once the step count is low.

Aug 1, 2026

Deep dive

Why Image Generation Takes Fifty Steps

generative

diffusion

A diffusion model runs the same network dozens of times to make one picture. This post asks why it cannot run it once — and answers with an experiment where the network is provably perfect and the sampler still fails completely.

Aug 1, 2026

Deep dive

The Noise Schedule Decides Where Your Timesteps Land

generative

diffusion

Both previous posts took the linear schedule from the DDPM paper without comment. It decides how a fixed budget of timesteps is distributed across signal-to-noise ratios — and the linear and cosine schedules distribute it very differently, which is one reason cosine was introduced and often behaves differently from the original linear schedule.

Jul 31, 2026

Note

Classifier-Free Guidance Is Extrapolation, Not Interpolation

generative

diffusion

CFG computes what the model predicts with your prompt and without it, then pushes past the conditional prediction rather than blending toward it. That’s why turning the knob up keeps working — until it very visibly doesn’t, with no error and no warning.

Jul 30, 2026

Note

Diffusion from Scratch: One Model, Two Samplers

generative

diffusion

A diffusion model is trained to predict the noise that was added to an image, not the image. That parameterisation gives a fixed-scale training target — and once the network is trained, DDPM and DDIM use that same denoiser to trace different reverse trajectories through the same noise levels.

Jul 29, 2026

Deep dive