Trained on the official manhua and it's should be as close to the original as possible, i "stress test" this LoRA and it really consistent in both SFW and NSFW variants.

How to Use It Effectively

Trigger Word: driftpixel_YoonKarma

⚠️ Important: Do not use the trigger word by itself — results will be half-baked. Always combine it with detailed descriptions of pose, expression, outfit, lighting, and mood for the best quality.

Recommended Settings

• LoRA Weight: 0.9 – 1.2

• Sampler: Euler a (recommended) Alternative: DPM++ 2M Karras

• Scheduler: Normal or Beta 1.1

• Steps: 20 – 40

• CFG Scale: 5 – 7

• Resolution: 768×1024 (great starting point) I often use 832×1216

Recommended Universal Negative Prompt:

low quality, worst quality, blurry, jpeg artifacts, bad anatomy, bad hands, deformed, distorted, disfigured, extra limbs, missing limbs, malformed limbs, fused fingers, too many fingers, mutated hands, poorly drawn hands, poorly drawn face, long neck, gross proportions, cropped, out of frame, watermark, text, signature, logo, username

Bonus for ComfyUI Users

A full ComfyUI workflow is embedded in the metadata of every sample image I uploaded. Just download any showcase image and drag & drop it into ComfyUI — it will load the complete workflow automatically.

This LoRA represents many hours of careful training and testing. If you find it useful, please 👍 like, 💬 leave a comment with your results, and feel free to ⚡ tip. it really helps support future projects!