Verified: 8 months ago
SafeTensor
Type | |
Stats | 900 11,780 |
Reviews | (195) |
Published | Aug 8, 2024 |
Base Model | |
Training | Steps: 2,050 |
Usage Tips | Clip Skip: 2 Strength: 1 |
Hash | AutoV2 98B285D9D1 |
SDXL LoRA/LoCon for Pony V6/Pony merges, trained on Eatsleep1111's style. Support the artist: Eatsleep1111.
Works well at 1 weight.
What I use & settings: Euler A Sampler & Align Your Steps Scheduler - 50 Steps, Base Res of 832x1216 with 2x Hires fix, 10 Hires steps & 0.3 denoise, ADetailer at 1024x1024 & 0.3 Denoise, (PAG) Perturbed-Attention Guidance: 4 PAG Scale | 1 Rescale Pag | Rescale Mode "Full", 5 CFG w/ PAG Rescale.
Using reForge by Panchovix: ReForge