New Chinese Opera Style Hanfu

0
0
0
0
PhotographyGirlObject Enhance
Cập nhật gần đây: Đăng tải lần đầu:
Photography,Girl,Object Enhance,LoRA,FLUX.1Image info
Photography,Girl,Object Enhance,LoRA,FLUX.1Image info
Photography,Girl,Object Enhance,LoRA,FLUX.1Image info
Photography,Girl,Object Enhance,LoRA,FLUX.1Image info
Photography,Girl,Object Enhance,LoRA,FLUX.1Image info
Photography,Girl,Object Enhance,LoRA,FLUX.1Image info

Opera Style Revision

Compatible with the local FP8 version, backward compatible, sampling euler normal, 25 steps, increase steps for bad hands.

Recommended image size is 1024X1536, large images are less likely to have bad hands, online live images can enable high-definition repair, full-body images are also less likely to have bad hands, if vertical stripes appear, disable high-definition repair.

Recommended Lora weight is 0.7 or below, CFG guidance 3.5, do not apply local live image information automatically.

Example of prompt: Reference example image, pay attention to CFG guidance.

If you don't know how to write it, use reverse engineering, preferably in natural language, recommend Florence2 or joy2, combine with Lora and write casually except for "1girl", clothing words should not be too different from Lora style.

Works well when fused with portrait or other types of Lora, explore weight ratios yourself, collisions between Loras will produce interesting effects.

Although Lora is good, don't overdo it.

Return many creative beautiful images, everyone!

Looking forward to your return images and suggestions; Penguin Communication Group: 439197807                                                                                                        

Thảo luận

Phổ biến nhất
|
Mới nhất
Gửi
Đến sớm
Tải xuống
(0.00KB)
Chi tiết
Loại
Số lần tạo hình ảnh trực tuyến
0
Tải xuống
0
Tham Số Đề Xuất
Kết Hợp Đề XuấtCheckpoint
Không
Cân nặng được khuyến nghị.
0.7
CFG
3.5
VAE
Không

Bộ sưu tập hình ảnh

Phổ biến nhất
|
Mới nhất