Hello everyone! Today, I'm uploading some of the requested models. I'll start with Nanami Konoe!<hr />Prompt: look at "Trigger Words" on the right side 👉Changelog: look at "About this Version" on the right side 👉<h1></h1>My Settings (forge neo)*:*note, the example images only use high-res fix.Checkpoint: waiNSFWIllustrious_v170 for IL, waiANIMA_v10Base10 for animaSampler: Euler a for both IL and AnimaScheduler: <s>Automatic</s> Karras for IL, Normal for AnimaSamplings Steps: 30 for both IL and AnimaHires.fix: R-ERSGAN 4x+Anime6B, 20 Hires steps for both IL and Anima, Denoising Strength: 0.3-0.5 for IL and 0.2-0.4 for animaExtensions (nice to have for little more Details):Adetailer (highly recommended)

Nanami Konoe - Lamune

Nanami Konoe ver 1.01 - Lamune-step00000800.safetensors

Nanami Konoe v1.0

Hello everyone! Today, I'm uploading some of the requested models from the anime Lamune. I'll start with Nanami Konoe! A small note: Since the model includes many different outfits, some of them may differ slightly from their appearance in the original anime. A little technical info (for those interested):I've changed my Chainer upscaling workflow for low-quality anime screenshots. Previously, the images sometimes looked washed out after upscaling with my usual method (4× supersampling followed by resizing to 50% of the original upscaled dimensions). Instead, I now resize the anime screenshots before upscaling them by 4×. This approach seems to preserve line quality better and produces less washed-out results. The trade-off is that resizing the images beforehand can theoretically remove some fine details. However, for low-quality screenshots, the overall results seem to be better with this approach. Thanks for your patience, all your requests, and your continued support. I really appreciate it! ❤️<hr />Prompt: look at "Trigger Words" on the right side 👉Changelog: look at "About this Version" on the right side 👉<h1></h1>My Settings (forge neo)*:*note, the example images only use high-res fix.Checkpoint: waiNSFWIllustrious_v170 for IL, waiANIMA_v10Base10 for animaSampler: Euler a for both IL and AnimaScheduler: <s>Automatic</s> Karras for IL, Normal for AnimaSamplings Steps: 30 for both IL and AnimaHires.fix: R-ERSGAN 4x+Anime6B, 20 Hires steps for both IL and Anima, Denoising Strength: 0.3-0.5 for IL and 0.2-0.4 for animaExtensions (nice to have for little more Details):Adetailer (highly recommended)

Files with SHA256: 46e9ff358bd28b0d9b199180d1bbce765d990e872093f455f0ec048360ed3d92