Switch

AlignYourStepsScheduler

SamplerCustom

KSamplerSelect

VAE Decode

ApplyRAUNet

Automatic CFG

Load VAE

Apply ELLA

Load ELLA Model

ella-sd1.5-tsc-t5xl.safetensors

T5 Text Encode #ELLA

Load T5 TextEncoder #ELLA

flan-t5-xl-encoder-only-bf16

Combine CLIP & ELLA Embeds

Text Box

upper body portrait of 1girl wear a black color turtleneck sweater,A proud and confident expression,long hair,look at viewers,studio fashion portrait,studio light,pure white background, solo


embedding:EasyNegativeV2.safetensors embedding:ng_deepnegative_v1_64t.pt, <(worst quality, normal quality, low quality:2.0), (paintings, sketches, 3d face:1.3)>, freckles, lowres, monochrome, grayscale, bad proportions, nsfw, sandals,

Save Image

Load Checkpoint

ApplyMSWMSAAttention

Empty Latent Image

CLIP Text Encode (Prompt)

cond

embeds

ELLA_EMBEDS

clip

text

CONDITIONING

text_encoder

STRING

T5_TEXT_ENCODER

input1

input2

input3

select

selected_label

selected_index

ELLA

model

ella

positive

negative

sigmas

MODEL

Ella or Combine

sampler

latent_image

output

denoised_output

SAMPLER

SIGMAS

LATENT

samples

IMAGE

images

CLIP

Datou

<p>original author: https://pastebin.com/yvGvxSPr</p>

SD1.5 ELLA + HI-DIFFUSION

OpenArt

Workflows

Active Sessions

SD1.5 ELLA + HI-DIFFUSION

Description

Discussion

Loading...