AIイラストおじさん - illustrious V2.1
Recommended Prompts
score_9, score_8_up, score_7_up, score_6_up, 1girl, breasts, long hair, large breasts, blush, looking at viewer, solo
Recommended Negative Prompts
deformities, simple background, mosaic censoring, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, censored, worst quality, low quality, quality bad, hands bad, eyes bad, face bad, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, monochrome, greyscale
Recommended Parameters
samplers
steps
cfg
clip skip
resolution
Recommended Hires (High-Resolution) Parameters
upscaler
Tips
For a well-defined image, request Priorit - high +10 or higher.
Adding many LoRAs may cause instability due to high definition and configurations.
Spending more Buzz is recommended for excellent high resolution images.
If you find faults or have doubts, report immediately.
Version Highlights
Please note: Avoid setting this LoRA to Strength 1.0 when using multiple LoRAs simultaneously. The optimal range is typically between 0.05 and 0.7.
NEGATIVES PROMPTS: censored, deformities, simple background, mosaic censoring, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, quality bad, hands bad, eyes bad, face bad, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, monochrome, greyscale
main compatibility: WAI-NSFW-illustrious-SDXL and NoobAI-XL (NAI-XL)
Recommended CFG Scale: 3.0 - 6.0
Recommended sampler: Euler a for softer tone and DPM++ 2M Karras for stronger or heavier tones.
Steps will depend: 24 - 50 LoRA tries to maintain quality.
recommendations for a well-defined image: request Priorit - high +10 or higher.
upscale: the recommended option is Hi-res fix - between - 0.35 and 0.55 (higher ones will make the image more diverse and strange), second option if the image is good (if the eyes are not deformed), Face fix - between - 0.45 and 0.80
parameters to be placed:
In case of adding many LoRAs, it is believed that this LoRA will become unstable due to its high definition and configurations.
This LoRA generates high resolution images, so spending more Buzz is the best option for an excellent image!
If you have any doubts or find faults, report them immediately!
Contributor
Model Details
Discussion
Please log in to leave a comment.