A barebones repository for my base models. Using adetailer is strongly recommended. You can find face detection models here: https://arcenciel.io/models/13770

Illustrious XL Series

MonsterCoffeeC4!

Base Model: IllustriousXL v0.1 (SDXL 1.0)
CivitAI Link
Description: Focuses on semirealism.
Recommended Settings
- Positive Prompt: masterpiece, best quality, realistic
- Negative Prompt: worst quality, low quality, jpeg artifacts, multiple views, 4koma, bad anatomy, bad hands, text
- Sampler: Euler A Align Your Steps or DPM++ 2M SGM Uniform, 24 steps
- Resolution: 832x1216 (Portrait) / 1216x832 (Landscape) / 1024x1024 (Square)
- CFG Scale: 6
- Hires: 2x-AnimeSharpV3, Upscaled by 1.5×, Denoise 0.3, 12 steps
Models Merged
- Madly Mix vol. IllustriousXL v2.0
- Rillusm - Realistic Illustrious v1.1
- Sade Abyss Style GLoRA
- Zumidraws Style GLoRA

MonsterCoffeeVenom!

Base Model: IllustriousXL v0.1 (SDXL 1.0)
Arc en Ciel Link
Description: Focuses on semirealism. Has a tendency to generate foreheads, likely having to do with the Rillusm checkpoint that was merged in.
Recommended Settings
- Positive Prompt: masterpiece, best quality, realistic
- Negative Prompt (at minimum): worst quality, low quality
- Sampler: Euler A Align Your Steps, 24 steps
- Resolution: 832x1216 (Portrait) / 1216x832 (Landscape) / 1024x1024 (Square)
- CFG Scale: 6
- Hires: 2x-AnimeSharpV3, Upscaled by 1.5×, Denoise 0.3, 12 steps
Models Merged
- NoobAI-XL v1.1 Epsilon Pred
- Rillusm - Realistic Illustrious v2.0
- Sade Abyss Style GLoRA
- Zumidraws Style GLoRA

MonsterCoffee5HourEnergy!

Base Model: Chenkin Noob XL v0.1 (SDXL 1.0)
CivitAI Link, Mirror
Description: A blend comprising of semirealistic art styles. A VAE is already baked in.
Recommended Settings
- Positive Prompt: realistic, nose, lips
- Negative Prompt (at minimum): worst quality, low quality
- Sampler: Euler A Align Your Steps, 24 steps
- Resolution: 832x1216 (Portrait) / 1216x832 (Landscape) / 1024x1024 (Square)
- CFG Scale: 6
- Hires: 2x-AnimeSharpV3, Upscaled by 1.5×, Denoise Strength: 0.3, 12 hires steps
Models Merged
- Chenkin Noob XL v0.1
- Dandon Fuga Style GLoRA
- Sciamano240 Style GLoRA
- NeoArtCore Style GLoRA
- Zumidraws Style GLoRA

MonsterCoffeeGhost!

Base Model: Chenkin Noob XL v0.1 (SDXL 1.0)
CivitAI Link, Mirror
Description: A blend with a stylistic approach to 2D. A VAE is already baked in.
Recommended Settings
- Positive Prompt: masterpiece, best quality
- Negative Prompt (at minimum): worst quality, low quality
- Sampler: Euler A Align Your Steps, 24 steps
- Resolution: 832x1216 (Portrait) / 1216x832 (Landscape) / 1024x1024 (Square)
- CFG Scale: 6
- Hires: 2x-AnimeSharpV4, Upscaled by 1.5×, Denoise Strength: 0.3, 12 hires steps
Models Merged
- Chenkin Noob XL v0.1
- MLeth Style GLoRA
- Tsuaii Style GLoRA
- Kajinman Style GLoRA
- Yatagarasupen Style GLoRA

ChenkinNoob-XL Series

Caffeinated Sweets Series: Starburst

If you use ComfyUI, use the ModelSamplingSD3 node. If not, ensure your webUI such as reForge supports advanced model sampling mode and the adjustment of shift. Instructions can be found in ChenkinNoob-XL's repo.
Base Model: ChenkinNoob-XL v0.3 Rectified Flow (SDXL 1.0)
CivitAI Link, Mirror
Recommended Settings
- Positive Prompt: masterpiece, best quality, newest
- Negative Prompt: worst quality, low quality, jpeg artifacts, multiple views, 4koma, bad anatomy, bad hands, text
- Sampler: Euler A Comfy RF Normal, 24 steps
- Resolution: 832x1216 (Portrait) / 1216x832 (Landscape) / 1024x1024 (Square)
- CFG Scale: 5-6
- Hires: 2x-AnimeSharpV4, Upscaled by 1.5×, Denoise 0.15, 10 steps
- Shift: 6.0
Models Merged
- Chenkin Noob XL v0.3 RF
- Kajinman Style LoCon
- minicop2001 Style LoCon
- frostedink Style LoCon
- HotVR Style LoCon
- roropull Style LoCon