Anim8Draw is a specialized model that generates anime and manga-style artwork.
Thanks for trying this model! 🤗 I hope it meets your creative needs and brings you joy. And don't forget to give a 👍 if you like it! ;>
If you'd like to use my checkpoint for online generation, please visit here:
👉 SeaArt AI | Tensor.Art | ShakkerAI (No longer updated)
Beta version:
+ Hand anatomy issues: The version may produce anatomically incorrect or distorted hands.
+ Results may vary and require additional post-processing
+ Recommend using Upscale and Adetailer for face details
Recommended Setting Examples:
Clip skip: 2
Recommended resolution: 1024x1024, 832x1216. 864x1152, 896x1152, 1216x832, 1152x864, 1152x896, 1024x1630
Sampler: DPM++ 2M Karras, DPM++ 2M, Euler a,
Steps: 20~40
CFG: 4~7
Recommend using Hires Fix.
Use Adetailer for face details
Feel free to share results!
Please DO NOT re-upload my model to other platforms.
Description
Anim8Draw v4.0 Changelog:
Sharper lineart
Better hair detail
Improved eyes
Better shading
Cleaner backgrounds
FAQ
Comments (3)
-199，评价是v4融合了v3和v2，结果是既有v2奇怪的水痕的缺点，又有v3人体的崩坏。也不知道是不是我画师串和画风LoRA的问题，至少我的测试不如v2
So far, I’ve used every version, including the Early Access 4.0. Each version has its own unique features, but I feel that version 3.1 is currently heading in the right direction. I think it strikes just the right balance, delivering excellent results for a wide range of creative styles. I’d definitely go with version 3.1.
Bro, what kind of mysterious material did you add to this model? When I try running some of my strongly personal-style LoRA models on other Checkpoint models, the generated images often come out incomplete or damaged, as if the LoRA style has overly eroded and taken over everything. But surprisingly, in your model, this problem doesn't exist at all. What kind of magic is this?