NONAMEmix-Vpred

NOTICE

• This is a v-prediction model.
• No longer includes noob AI.
• 99% of this checkpoint is made with RouWei.

Model Description

• Model type: Diffusion-based text-to-image generative model
• Model prompt style: Booru-tags
• License: Fair AI Public License 1.0-SD

Merge Recipe

• RouWei 0.7 vpred (A)
  • (https://civitai.com/models/950531/rouwei)
• CyberRealistic XL (B)
  • (https://civitai.com/models/312530/cyberrealistic-xl)
• stable-diffusion-xl-base-1.0 (C)
  • (https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0)
• SPO-SDXL_4k-p_10ep_LoRA_webui (D)
  • (https://civitai.com/models/510261/spo-sdxl4k-p10eplorawebui)
• My LoRA (E)
  • Trained by 300 images AI-generated

A + (B - C) * 0.5 = F
F + D + E * 0.5 = This Model v2.0

Recommended settings

• Positive prompts:

masterpiece, best quality

• Negative prompts:

worst quality, low quality, watermark

• CFG: 3-5
• Sampling steps: 28
• Sampler: Euler a
• Supported Resolution:

1024 x 1024, 1152 x 896, 896 x 1152, 1216 x 832, 832 x 1216, 1344 x 768, 768 x 1344, 1536 x 640, 640 x 1536

Hires.fix Setting

• Upscaler: 4x-UltraSharp
• Hires step: 10-15
• Denoising: 0.3-0.4

Thanks

• First of all, to Minthy for RouWei. That has the best prompt following I know of!
• And, of course, thanks for the Animagine. Animagine was the first signpost that led me to use SDXL, and it continues to do so!