StableDiffusionWebUI: Sampling Steps

Preface

This section introduces the sampling steps in sdwebui.

Sampling Steps

The previous section covered samplers: https://blog.vincentqiao.com/sd-sampler

Sampling steps refer to the number of noise addition and denoising iterations during image generation.

Prompt - Chinese

一个男孩，坐在椅子上，上半身，靠近窗口，蓝色上衣，黑色头发，戴眼镜，面向镜头，左手边是窗子

Prompt - English

A boy sitting on a chair, upper body, close to the window, blue shirt, black hair, wearing glasses, facing the camera, with the window on his left side

Checkpoint

Samaritan 3d Cartoon

Seed

3255625742

Sampler

Euler a

Steps - 1

Steps - 5

Steps - 10

Steps - 15

Steps - 20

Steps - 25

Steps - 30

Steps - 50

Steps - 100

Summary

Too few steps means the denoising process is incomplete, making the result basically unusable.
Too many steps yields diminishing returns in optimization.
Checkpoint authors usually recommend a specific step count. In practice, 20-30 steps generally produce good results.

SDWebUI

|—URL: https://sdwebui.ai/?from=vq

|—Authentic Stable Diffusion WebUI V1.10.1~

|—Top-tier 4090 GPU, premium members get exclusive single 4090 for image generation~

|—Supports Mac client, Windows client, and Web online usage~

|—Innovative instant model transfer, upload local models instantly~

|—Supports txt2img, img2img, extras, and image info

|—Supports CLIP interrogation and DeepBooru interrogation

|—Supports HD upscaling (Hires.fix), with 21 built-in upscaling models

|—Supports detail restoration (ADetailer), with 11 built-in restoration models

|—Supports ControlNet, with 44 built-in preprocessors and 68 built-in models

|—Supports ultra-HD upscaling (Tiled Diffusion, Demo Fusion, Tiled VAE)

|—Supports old photo restoration, with 21 built-in upscaling models, plus GFPGAN and CodeFormer restoration models

|—Built-in 2 common Styles, 11 popular Checkpoint models, 7 popular LoRA models, 14 common embeddings

|—Built-in oldsix prompt plugin

|—Built-in 23-lesson beginner tutorial