# IDK Eye Shapes V2

## About

LoRA focused on the 'Eye Shapes' concept for Illustrious-based SDXL models. It strengthens generation of glowing eyes.

PT-BR: LoRA focada no conceito 'Eye Shapes' para modelos SDXL baseados em Illustrious. Reforca a geracao de glowing eyes.

## Trigger words

### Primary triggers

```

glowing eyes

```

### Supporting tags

Frequent thematic tags from the training set - combine them with the primary triggers to steer the concept:

```

glowing, red eyes, blue eyes, green eyes, yellow eyes, male focus, weapon, long hair, armor, short hair, blonde hair, no humans, fire, sword

```

## Example prompts

```

masterpiece, best quality, absurdres, close-up portrait, woman with almond shaped eyes, expressive gaze, detailed iris, soft makeup, <lora:IDK_eye-shapes_V2:0.9>

```

```

masterpiece, best quality, absurdres, close-up portrait, woman with large doe eyes, seductive bedroom eyes, detailed iris, dramatic eyeliner, <lora:IDK_eye-shapes_V2:0.9>

```

```

masterpiece, best quality, absurdres, close-up portrait, woman with cat-like upturned eyes, expressive gaze, detailed iris, soft makeup, <lora:IDK_eye-shapes_V2:0.9>

```

## Preview generation settings

| Setting | Value |

| --- | --- |

| Checkpoint | IDN_Illustrious_V10_T |

| Size | 1280x720 + hires 1.5x RealESRGAN 4x+) = 1920x1080 |

| Sampler | Euler a |

| Steps | 28 (+14 hires) |

| CFG scale | 5.5 |

| Denoising strength | 0.4 |

| Seeds | 1000-1009 |

Negative prompt: worst quality, low quality, bad anatomy, bad hands, extra digits, fewer digits, missing fingers, jpeg artifacts, signature, watermark, username, blurry, lowres, bad proportions, deformed, mutated

## Recommended weight

Use 0.7-1.0. Start at 0.9 and adjust to taste.

## Changelog V2

- New, larger and more diverse training dataset.

- 1080p preview images.

Base model: Illustrious SDXL IDN_Illustrious_V10_T)

Training pipeline: [IDK-Trainer](https://github.com/NOTcisLol/IDK-Trainer)

File: IDK_eye-shapes_V2.safetensors