AstolfoXL (AK NIL1.5 v1.2 based, 255c based)

AK NIL1.5 v1.2 based

Base model: AstolfoKarmix-XL, version NIL1.5 v1.2
All training settings remains the same. Suggested to not use directly.
48 days for 1EP. Train 1 EP only. Save per 11274 steps. 11274x69=777917-11

Coming soon

255c based

Confirmed the base model had wrong merging recipe. Suggested to not use directly.
Base model: AstolfoMix-XL, version 255c
Tech report: ch06
Dataset (images > latents): danbooru2024, e621_2024
Dataset (tags + captions): meta_lat.json
1 step = 16 images, 4x RTX 3090 24G.
75-100+ days for 1EP. Train 1 EP only. Save per 10k steps.
Core concept: "Unsupervised learning". Expectation: MID

parameters

anthro, furry, astolfo
Steps: 64, Sampler: Euler, Schedule type: Automatic, CFG scale: 4, Seed: 2206138433, Size: 1024x1024, Model hash: 792e768cd2, Model: 1ep-AstolfoXL, VAE hash: 235745af8d, VAE: sdxl-vae-fp16-fix.vae.safetensors, Clip skip: 2, SEG Active: True, SEG Blur Sigma: 11, SEG Start Step: 0, SEG End Step: 2048, PAG Active: True, PAG SANF: True, PAG Scale: 1, PAG Start Step: 0, PAG End Step: 2048, Version: v1.10.1

Downloads last month: 198