Load VAE

ModelSamplingSD3

KSampler (Advanced)

CLIP Text Encode (Positive Prompt)

LoraLoaderModelOnly

CLIP Text Encode (Negative Prompt)

色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走

Save Image

Load CLIP

VAE Decode

EmptySD3LatentImage

Load Diffusion Model

Zane Image Directory Loader

Zane Prompt Expand OpenAI

Show Text 🐍

Fine art hyper realistic photography of a Chinese woman with short blonde hair, wearing a white, semi-sheer, ruffled blouse with voluminous sleeves and beige high-waisted pants with a black belt, standing with one hand in her pocket. The image is A8, 8K, masterpiece, high definition, best quality, suggestive, a bit nsfw. The woman’s expression is calm and slightly introspective, with a soft, natural look on her face. The background is plain white and blurred, creating a shallow depth of field that emphasizes the subject. The mood is serene and subtly provocative, with an extremely high aesthetic quality. The image is portrait orientation.

Preview Image

🔧 Image Resize

Anything Everywhere

Step1 - Load models

MODEL

Model Links

text

STRING

IMAGE

MASK

image

width

height

LATENT

LatentInput

model

PromptPositive

anything2

anything3

anything11

CLIP

wan_2.1_vae.safetensors

prompt

ImageReference

clip

CONDITIONING

positive

negative

latent_image

images

samples

人生在国贸

wan2.2-t2i-repaint-r2

OpenArt

Workflows

Active Sessions

wan2.2-t2i-repaint-r2

Description

Discussion

Loading...