New Spageta Style

Information for the Pony version is coming soon.

Hello CivitAI community!

This is my first LoRA release, and I put a lot of effort into creating it. I've meticulously removed nearly all watermarks, signatures, and artist names (approximately 99.5%) from the training dataset. Therefore, you don't need to include
negative prompts like "watermark" or "signature" and "artist name" when using this LoRA. Enjoy!

*I used ChatGPT to help translate this description. I apologize if anything is unclear.

To get the most out of this LoRA, I recommend the following:

• Based on my testing, I can confidently recommend these Checkpoints:
  WAI-NSFW-Illustrious and Hassaku-XL-Illustrious. They deliver excellent results.

• LoRA Strength (Weight): 0.75 to 1.0 (best range for capturing the artist's signature style.)

• Sampling Method 1: DPM++ 2M SDE Karras: 20+ Sampling Steps

• Sampling Method 2: Euler A: 25+ Sampling Steps

• CFG Scale: 2 to 5.5 (to 7 for Euler A)

• Clip Skip: 2 (Not required for ComfyUI)

• Resolution: Portrait - 832x1216, 1024x1024, 1024x1360, 1536x1536, 1248x1824(BEST)

  Landscape - 1216x832, 1360x1024

Highly recommended

• 2x Upscale with Remacri.

• Adetailer or FaceDetailer(ComfyUI) for fix Face.

Recommended Tags:

Triggers Word:

spagetastyle,

Comic Panels:

comic, text, speech bubble, 4koma,

Movement:

motion lines, onomatopoeia,

Sound Effects:

sound effects, hearts, haa, hah, ooouch!, hmph!, smack!, oh, ooh, ah!, mngh! nhg!, slorp, mmm, bdmp, glup, plap, ugh!, slap,

If you'd like to support the artist, check this out:

X ( Twitter) - Patreon