Overview

This guide explains how to generate images similar to mine using my Anima models on Forge Neo.

⚙️ Core Settings

Steps: 30

Sampler: Euler a
Scheduler: Normal

CFG Scale: 5

Shift: 3

🧩 Model & Precision

Model/Checkpoint: WAI-ANIMA
Diffusion in Low Bits: Automatic

VAE: qwen_image_vae.safetensors

Text Encoder: qwen_3_06b_base.safetensors

👤 ADetailer – Face (2nd Pass)

Detector Model: face_yolov8n.pt
Version: aadetailer-neoforge

⚠️ Important Note

I use the default settings

Never apply ADetailer on the first pass.
Always run ADetailer after the base generation to avoid unintended changes to:

• Facial identity

• Eye color

• Expression

• Proportions

This is especially important in Forge neo to preserve consistency.

🖼️ Hires.fix (Optional)

Hires Upscale: 1.5

Hires Steps: 20

Hires Upscaler: R-ESRGAN 4x+ Anime6B

Denoising Strength: 0.3 ~ 0.5

Hires CFG Scale: 0.3 ~ 0.5

Shift: 0.3

⚠️ Be aware that adding this stage the generation time will increase a bit.

🚫 Negative Prompt Recommendations

NO NSFW Negative Prompt

((blush, sweat, sweatdrop, dark, night)), worst quality, low quality, bad quality, worst detail, blurry, jpeg artifacts, lowres, sketch:1.2, score_1, score_2, score_3, censor, ((watermark, text, logo, signature, artist name, name, labels, artist signature)), safe, sensitive, nsfw, explicit

NSFW Negative Prompt

worst quality, low quality, bad quality, worst detail, blurry, jpeg artifacts, lowres, sketch:1.2, score_1, score_2, score_3, censor, ((watermark, text, logo, signature, artist name, name, labels, artist signature))

📌 Additional Notes & Tips

This is not a final or exhaustive guide, please feel free to experiment and share suggestions to help improve it. Thank you.