🎨 Rayne's Feef/Feefal Style — SDXL LoRA
A style LoRA trained to capture the distinctive visual qualities associated with Feefal's artwork — expressive characters, stylized features, atmospheric rendering, and a beautifully illustrative finish.
This LoRA is intended for SDXL 1.0 and works well for character illustrations, portraits, fantasy subjects, and other artwork where you want the trained style to come through.
✨ Trigger Word
Use:
feefal_style
in your prompt to help activate the trained style.
Example
feefal_style, portrait of a fantasy girl, expressive eyes, flowing hair, soft atmospheric lighting, detailed illustration
You can combine the trigger with your usual character, clothing, composition, lighting, and environment tags.
⚙️ Recommended Use
Base model: SDXL 1.0
Trigger: feefal_style
For LoRA strength, experiment based on the checkpoint and prompt you're using. A good starting range is around 0.7–1.0, then adjust downward if the style becomes too dominant or upward if you want it to have a stronger influence.
Different SDXL checkpoints may interpret the LoRA slightly differently, so experimentation is encouraged.
💡 Prompting Tips
For stronger results, try describing the subject and scene clearly instead of relying entirely on the LoRA.
Things that pair nicely with the style include:
Character portraits
Fantasy characters
Expressive faces
Flowing hair
Atmospheric or dramatic lighting
Painterly backgrounds
Detailed clothing and accessories
Soft or moody color palettes
Illustrative compositions
The LoRA can also be mixed with other concepts and character LoRAs, though reducing individual LoRA strengths may help when combining several models.
🖌️ Artist Credit & Disclosure
This is an unofficial, fan-made style LoRA and is not created, endorsed, sponsored, or affiliated with the original artist.
The artist whose work was used as the stylistic training reference is Feefal.
Please check out and support the original artist here:
❤️ Feefal on X / Twitter
If you enjoy the aesthetic this model is based on, please consider following and supporting the artist and their original work.
📌 Notes
This LoRA is a tool for creating new AI-generated images influenced by characteristics learned during training. Results will vary depending on the checkpoint, sampler, resolution, prompt, seed, and LoRA strength.
Feel free to share your generations — I'd love to see what people make with it. ✨
Have fun experimenting!



