【Meow Ka_Tifa】(Unique Edition)

0
0
0
0
CGI styleAnime CharacterGirlCharacter
Cập nhật gần đây: Đăng tải lần đầu:
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info
CGI style,Anime Character,Girl,LoRA,SD 1.5Image info

Welcome to use this Tifa Lora model. This version of Tifa is based on the official settings of the Final Fantasy 7 Remake Red-Eyed Tifa, which is the most compliant with the original version. This Tifa Lora is the first character I have created a model for. Any suggestions for this model are welcome, feel free to provide feedback, and I will follow up in the subsequent versions. This model is compatible with 2D, 2.5D, and realistic base models.

Model Feedback AI Technology Exchange QQ Group: 377209248

Suggested Drawing Parameters

LORA Drawing Weight Recommendation: 0.8~1 - Face won't resemble below 0.8 weight. I have tested the default weight of 1 on many base models and the model does not crash. In case of base model compatibility issues, please try reducing the weight by 0.05 for experimentation.

Recommended Tags: 1girl, Tifa Lockhart, solo, thigh-highs, skirt, suspenders, long hair, crop top, gloves, red footwear, suspender skirt, black thigh-highs, fingerless gloves, full body, black top, black skirt, midriff, navel, elbow gloves, boots, black hair, white top, clenched hands, red eyes, fighting stance, brown hair, full body

Recommended Parameters:

Sampling Method:

Euler a: 20~30 steps.

DPM++ 2S a Karras: 25~30 steps.

DPM++ SDE Karras:20~30 steps.

DPM++ 2M SDE Karras:25~30 steps.

DDIM:30 steps and above.

Prompt Word Guidance Coefficient (CFG Scale):7

Character Portrait Resolution:512x768, 768x768, 512x1024

High-Resolution Restoration (Hires.fix):

Scaling Algorithm:R-ESRGAN 4x+Anime6b, 4x-UltraSharp, 4x_SmolFace_200k

High-Resolution Iteration Steps:10~15

Redraw Range:0.3~0.5

Clip Skip:2

Suggested OpenADetailer

Negatives:(worst quality, low quality: 2), monochrome, zombie, (interlocked fingers: 1.2)

This model does not contain any real human material image sources, and training code is sourced from the internet. The model is only used for scientific research and interest exchange.

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.
0.8
CFG
6
VAE
Không

Bộ sưu tập hình ảnh

Phổ biến nhất
|
Mới nhất