Miracle - Dream Space Master Film Style Simulation

0
0
0
0
RealisticStreet PhotographyGirlBoyObject Enhance
Cập nhật gần đây: Đăng tải lần đầu:
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info
Realistic,Street Photography,Girl,LoRA,SDXLImage info

**Miracle-injection Dream Space Master Film Simulation Lora Model SDXL Version**



**File Name:** Miracle-Dream Space Master Film Style Simulation_v20240301

**SafeTensors Trigger Word:** INJECTION (optional)

**Recommended Lora Weight:** 0.3 to 0.7, weight



The strengths and weaknesses of different movie styles change. Adding Master Element 0.6 is enough. 0.5 is the median value, and you can adjust Lora's weight yourself as needed.



(Weight 0.3 only adds cinematic feel, commonly used 0.5)

**Recommended sampling steps:** 30 to 50

**CFG scale:** 7 to 8

**Recommended sampling methods:** for realistic models, DPM++ 2M Karras, Euler a for drawing models, welcome to try other sampling methods; solutions such as **LCM Turbo** have not been tested.



**Recommended resolution:** The 6**40x1280 training set mainly focuses on the 1:2 ratio. It is suitable for mobile viewing. You can use 6**40x1280 to draw pictures. Overlay 1.5 times the image to add detail. The horizontal screen resolution with a redrawing amplitude of 0.4-0.6 can better simulate the movie feel, because the official bottom model is better at movie frames.



**Recommended settings for high-definition images:** 8x_nmkd-superscale_150000_g self-contained film background noise. Or use 4x-UltraSharp for the first layer, and NMKD for the second layer, recommend using large models: LEOSAM HelloWorld Shinsekai, SDS_FILM/Film Photography   Overlay using Lora:SDS_Film Grain_xl_slider to restore bottom noise. Welcome to try other big models, and thank you for the feedback. Please leave a message for testing questions.

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.
1
CFG
8
VAE
Không

Bộ sưu tập hình ảnh

Phổ biến nhất
|
Mới nhất