This style LoRA was trained using 212 pure illustration source images, excluding any anime materials.
For best results, use a weight of 0.7–1.0 when using this LoRA alone. When combining it with other style LoRAs, feel free to lower the weight as needed. This LoRA is great for enhancing the realism and texture of the skin.
Prompts
@betmillustrationstyleV2.5 ➡️ Trained with higher resolution and more steps, offering better style fidelity and clarity compared to 2.0. Meanwhile, any issues with character diversity can be offset by using character tags or character LoRAs.
V2.0 ➡️ Enhanced art style fidelity and improved LoRA versatility. This version boasts superior prompt adherence. However, due to higher dataset fitting and strong character similarities within the training data, specific hairstyles and hair colors may occasionally trigger fixed character appearances, slightly reducing feature diversity compared to V1.0.
V1.0➡️Trained using anima_baseV1
Description
FAQ
Comments (2)
Could I ask why I couldn't achieve the same style even though I used your workflow? The only thing I did differently was the Diffusion Model. I couldn't find miaomiao20harem20Anima.TZS8.safetensors like the one you used, so I replaced it with miaomiao12 and miaomiao15 instead. However, I still couldn't get the desired results. Btw thank you for wonderful lora.
This LoRA was tested on Miaomiao 12, which was selected for its superior stability in rendering hands and feet. However, due to the model's strong inherent style (despite being reduced in v12), it may impact the LoRA's output. For best results, consider using a base model, Yume, or other low-stylization models.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.