Raelina/Raejin-Anima
<style>
body { display: flex; align-items: center; justify-content: center; height: 100vh; margin: 0; font-family: Arial, sans-serif; background-color: #f4f4f9; overflow: auto; }
.container { display: flex; flex-direction: column; align-items: center; justify-content: center; width: 100%; padding: 20px; }
.title-container { display: flex; flex-direction: column; justify-content: center; align-items: center; padding: 1em; border-radius: 10px; }
.title { font-size: 3em; font-family: 'Montserrat', sans-serif; text-align: center; font-weight: bold; }
.title span { background: -webkit-linear-gradient(45deg, #FF9D3D, #AB886D, #CB80AB); -webkit-background-clip: text; -webkit-text-fill-color: transparent; }
.gallery { display: grid; grid-template-columns: repeat(4, 1fr); gap: 10px; }
.gallery img { width: 100%; height: auto; margin-top: 0px; margin-bottom: 0px; border-radius: 10px; box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2); transition: transform 0.3s; }
.gallery img:hover { transform: scale(1.05); }
.note { font-size: 1em; opacity: 50%; text-align: center; margin-top: 20px; color: #555; }
</style>
<div class="container"> <div class="title-container"> <div class="title"><span>Raejin Anima</span></div> </div> <div class="gallery"> <img src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/To-K1S6lKzVC60Nh8_J.png" alt="Image 1"> <img src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/Ku5fvIDEoEkfXaJ25Sjjk.png" alt="Image 2"> <img src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/qUpk0SF2BiKbVyt8vCQZX.png" alt="Image 3"> <img src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/K4p1kVmS33DBtC8M1tf-l.png" alt="Image 4"> </div> </div>
Overview
Raejin Anima is a specialized fine-tune of the Anima base model, meticulously crafted to elevate its visual fidelity. This model addresses specific limitations found in the original, aiming to deliver a more polished, aesthetic, and balanced output while preserving the core strengths of the base model.
Model Details
- Developed by: Raelina
- Model type: Diffusion-based text-to-image generative model
- Finetuned from: Anima
- Prompt style: Booru-style tags and Natural Language
Training Details
v0.1 (Experimental Alpha)
- Base Model: anima-preview2
- Method: Full fine-tuning using the kohya/sd-scripts trainer.
- Stability: LLM Adapter was disabled during training to mitigate catastrophic forgetting, ensuring the model retains its foundational knowledge while adopting new stylistic improvements.
- Dataset: Trained on a curated set of 2,2k high-quality images, utilizing a hybrid captioning strategy of both booru-style tags and natural language to enhance prompt adherence and stylistic depth.
Recommended settings
- Positive prompts:
masterpiece, best quality, absurdres- Negative prompts:
low quality, worst quality, jpeg artifacts, bad anatomy, bad hands, bad proportions, mutation, deformed, extra digits, fewer digits, missing arms, missing legs, signature, watermark, username- CFG: 4-5
- Sampling steps: 30-50
- Sampler: Euler a or ER SDE
- Scheduled type: Normal
- Standard Resolution:
832 x 1216, 1216 x 832, 1152 x 896, 896 x 1152, 1344 x 768, 768 x 1344, 1024 x 1024Acknowledgements
A special thanks to the community and the developers who made this project possible:
- Circlestone Labs: For providing the incredible open-source Anima base model.
- Kohya_ss & Contributors: For the comprehensive and powerful open-source training framework
Support
If you’d like to support this project or my other developments, you can buy me a coffee on Ko-Fi
