Third LoRA's is here! Trained on Chinese furry author: 灰黑糯米麻薯大福 (https://x.com/Yifeng_LZ)

Please support the real artists!

Based on Yiff in Hell model.

Works best with IVFF GRAYED model.

• Recommended CFG scale: 5–8

• Recommended steps: 24–44

• Trigger word: yflz

For best results, please use the negative prompt examples from the OP and include positive tags such as detailed fur, best quality, very aesthetic, high resolution, ultra‑detailed, and amazing quality. Please note that the LoRA may occasionally produce poorly rendered eyes.

UPD: new version is updated, total 55 epochs, less artifacts and bad rendered eyes.

Resolution limitations

This LoRA was trained on images with a native resolution of approximately 1024px on the shorter side, primarily using a 2:3 aspect ratio. For best results, use a similar aspect ratio and resolution as from the examples from OP.

Generating at resolutions below 1024px may result in:

• Blurriness and loss of fine details,

• Distorted anatomy and structural issues,

• Increased noise, artifacts, and overall degradation of image quality

Recommendations:

• For best results, generate at 1024x1536 or higher (e.g., 1024x1920 (2:3) or 1920x1080 (3:2)), this matches the training resolution and aspect ratio of the used training dataset, delivering the best detail and stability.

The default CivitAI generator and its built‑in workflow that using suboptimal settings (resolution handling) that hurt quality. It's better to use ComfyUI with the workflow provided below – it's based and fine‑tuned for this and other LoRA's.

Example workflow used in ComfyUI:

Be sure to check out the Suggested Resources tab too. Any tips, reviews, or feedback for future training would be appreciated!