AnimeBoysXL

AnimeBoysXL

Features

• ✔️ Good for inference: AnimeBoysXL is a flexible model which is good at generating images of anime boys and males-only content in a wide range of styles.

• ✔️ Good for training: AnimeBoysXL is suitable for further training, thanks to its neutral style and ability to recognize a great deal of concepts. Feel free to train your own anime boy model/LoRA from AnimeBoysXL.

• ❌ AnimeBoysXL is not optimized for creating anime girls. Please consider using other models for that purpose.

Inference Guide

• Prompt: Use tag-based prompts to describe your subject.

  • Append , best quality, amazing quality, best aesthetic, absurdres to the prompt to improve image quality.

  • (Optional) Append , year YYYY to the prompt to shift the output toward the prevalent style of that year. YYYY is a 4 digit year, e.g. , year 2023

• Negative prompt: Choose from one of the following two presets.

  1. Heavy (recommended): lowres, (bad:1.05), text, error, missing, extra, fewer, cropped, jpeg artifacts, worst quality, bad quality, watermark, bad aesthetic, unfinished, chromatic aberration, scan, scan artifacts, 1girl, breasts

  2. Light: lowres, jpeg artifacts, worst quality, watermark, blurry, bad aesthetic, 1girl, breasts

  • (Optional) Add , realistic, lips, nose to the negative prompt if you need a flat anime-like style face.

• VAE: Make sure you're using SDXL VAE.

• Sampling method, sampling steps and CFG scale: I find (Euler a, 28, 5) good. You are encouraged to experiment with other settings.

• Width and height: 832*1216 for portrait, 1024*1024 for square, and 1216*832 for landscape.

Training Details

AnimeBoysXL is trained from Stable Diffusion XL Base 1.0, on ~516k images.

The following tags are attached to the training data to make it easier to steer toward either more aesthetic or more flexible results.

Quality tags

• best quality: score >= 150

• amazing quality: score in the range of [100, 150)

• great quality: score in the range of [75, 100)

• normal quality: score in the range of [0, 75)

• bad quality: score in the range of (-5, 0)

• worst quality: score <= -5

Aesthetic tags

• best aesthetic: score >= 6.675

• great aesthetic: score in the range of [6.0, 6.675)

• normal aesthetic: score in the range of [5.0, 6.0)

• bad aesthetic: score < 5.0

Rating tags

• (None): general

• slightly nsfw: sensitive

• fairly nsfw: questionable

• very nsfw: explicit

Year tags

year YYYY where YYYY is in the range of [2005, 2023].

Training configurations

• Hardware: 4 * Nvidia A100 80GB GPUs

• Optimizer: AdaFactor

• Gradient accumulation steps: 8

• Batch size: 4 * 8 * 4 = 128

• Learning rates:

  • 8e-6 for U-Net

  • 5.2e-6 for text encoder 1 (CLIP ViT-L)

  • 4.8e-6 for text encoder 2 (OpenCLIP ViT-bigG)

• Learning rate schedule: constant with 250 warmup steps

• Mixed precision training type: BF16

• Epochs: 20