Description:
A soft oil-study look. Rendering turns painterly and semi-real, with gentle brushwork, muted warm tones and soft, diffused light. Fabrics like knit and fur get a tactile, painted texture. It feels like a quick oil portrait study.
Known quirk: it frames shots tighter (it can crop the top of the head) and drifts toward semi-realism.
This isn't a trained LoRA. No images were used to make it. It comes from steering neurons inside the base model:
An automated evolutionary search mixed and mutated groups of the UNet's feed-forward neurons; this look came out of round 2 as #443.
This file is six neuron groups across six layers, from early in the UNet through the middle block to the style layers.
The LoRA stores that push as a bias change, which is why the file is only about 28 KB.
How to use
No trigger word. Just load the LoRA; it acts on the model only, not the text encoder.
Strength 1 is the look in the previews. The effect scales linearly with strength: lower it for a lighter touch; negative values push the other way.
It stacks with other LoRAs.
Preview settings: 768×1024, 20 steps, CFG 4.5, DPM++ 2M Karras.
Compatibility
Made and tested in ComfyUI (Load LoRA / LoraLoaderModelOnly) on prefectIllustriousXL v4.0.
Other UIs: the file holds only bias changes, which some UIs (A1111 / Forge) may not load. If nothing changes, that's why.
Other Illustrious checkpoints: untested, but they share the base, so it may carry over.

