start quality prompts masterpiece, best quality, amazing quality, very aesthetic, newest, general,
negative prompts lowres, (bad), text, error, fewer, extra, missing, worst quality, jpeg artifacts, low quality, watermark, unfinished, displeasing, oldest, early, chromatic aberration, signature, extra digits, artistic error, username, scan, [abstract], multiple views, censored, bar censor, mosaic censoring, red pupils, colored inner hair,
v4.2
Version 4.2 was created by merging to improve stability when generating high-resolution images (1536x1536, 1728x1344, 1344x1728, 1824x1248, 1248x1824, 2016x1152, 1152x2016). While it performs better than ILXL 2.0 and previous versions, it's still not perfect—especially with wide horizontal images.
To prevent inconsistent art styles, the merge was also done to ensure a consistent and well-defined default style.
The recommended image resolutions are either the standard SDXL resolution with Hires fix at 1.5x, or high-res images around the 1536-pixel range.
• Quality tags, Rating
• Character
• Clothing, Actions
• Background, Surrounding objects, Lighting, etc.
• Artist, LoRA
Settings:
• Sampling: Euler a
• Schedule: Automatic
• Steps: 28
• CFG Scale: 5
• Width x Height: 832~1216 or 1152~2016(high-resolution)
When using Hires. fix:
• Upscaler: 4x-UltraSharp
• Steps: 10
• Denoising strength: 0.35
• Upscale by: 1.5
ADetailer: Only used for faces
v4.1
With the release of Illustrious-XL 2.0, I created this model as a test by merging it with some LoRAs and other models based on it.
I’m not sure if it’s due to the characteristics of Illustrious-XL 2.0, but there seems to be some instability in the outputs, so further testing will be necessary.
I do plan to attempt additional merges to improve stability, but since I intend to try various approaches, I’m not sure how long it will take.
• Quality tags, Rating
• Character
• Clothing, Actions
• Background, Surrounding objects, Lighting, etc.
• Artist, LoRA
Settings:
• Sampling