
Introducing a new SDXL LORA model inspired by Kaitlin Witcher (Instagram: @piddleass). This model aims to generate high-quality, painting-style portraits using the power of SDXL.
Key Features:
-
Built for SDXL, offering improved quality over SD1.5 versions
-
Specializes in creating artistic, painting-like portraits, however you can get some really good cinema photographic images with the right prompts
-
Developed using LEOSAM's HelloWorld XL as a base
Usage Tips:
-
Experiment with different settings to find the sweet spot for your desired results
-
Some generations produce excellent outcomes, however you do need to fine-tune. Recommended to read up on LEOSAM's prompt guidance
Development Insights:
-
This LORA is the result of multiple training sessions, settled on a dataset that focuses mainly on her higher quality portraits and selfies
-
Higher quality datasets may further improve consistency across generations
I encourage you to share your creations and optimal settings to help refine and improve the model's performance. Your feedback and experiments are valuable in helping us get better results.
Inspirations:
-
SD1.5 Kaitlin Witcher model by [pkmngotrnr]
-
LEOSAM's HelloWorld XL (https://civitai.com/models/43977?modelVersionId=570138)
Feel free to explore and unleash your creativity with this SDXL LORA model!
描述:
Trained on selection of Publicly available images from Instagram and Other
训练词语: KAITLINWITCHER
名称: KW_LORA.safetensors
大小 (KB): 1781944
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success