Type | |
Stats | 193 48 |
Reviews | (23) |
Published | Dec 25, 2024 |
Base Model | |
Training | Steps: 2,301 Epochs: 13 |
Usage Tips | Clip Skip: 1 Strength: 0.8 |
Trigger Words | EE_DA |
Hash | AutoV2 F344A2BF90 |
Trigger Word: EE_DA
Created from a curated dataset of 118 images, this LoRA aims to unify a broad spectrum of creative fantasy aesthetics, focusing on female characters. It was developed to explore richly detailed portraits and scenes featuring different fantasy themes, while maintaining visual consistency and flexibility in style.
Detail-Enhancing Terms:
Commonly repeated in the dataset are "highly detailed", "hyper-realistic", "intricate", "CGI-style", "digital painting" - including these will significantly improve clarity, texture, and overall quality.
Races & Proper Keywords:
To get the intended results for different fantasy archetypes, use the terms below exactly as written. They appear consistently in the dataset, so referencing them directly helps achieve more accurate outputs: "elf woman", "demoness", "fantasy woman".
My Settings:
LoRA Strength: 0.8
Model: Flux.1 Dev GGUF Q8_0
Text Encoder: fp8
Sampler: Euler
Scheduler: Beta
Steps: 20-30
Upscaler: Real-ESRGAN x4 Plus