If you want to show support, feel free to post, rate, follow, emote, or drop feedback! It helps!!! 😊🙇Most images in dataset are from somewhat recent to capture a specific style... Many of their images feature Mashima’s characters colored directly from his manga panels.<span style="background-color:rgb(26, 27, 30);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">No trigger word ... <span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">changes about this version<hr /><h1 id="illustrious-style-v2">Illustrious Style v2</h1>I used <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184?modelVersionId=2167369">WAI-illustrious-SDXL v15</a> for these images, feel free to try other checkpoints! Trained on IIllustrious XL 1.0 which is what the current WAI v15 is partly based on. Certain checkpoint or LoRA(s) may heavily influence the style.Generation details, prompts, and model info are listed with most images. CLIP IS always 2, displayed is bugged. CFG Scale 6 (5-7), 20-30 steps, and Euler a/DPM 2M as the sampler, Hires Remacri: Feel free to change!

Gaston18 Style (Mashima) [Pony/Illustrious]

No artist name used as a trigger word<ul><li>No trigger word required</li><li>Helps maintain the latest style</li><li>Reduces issues involving image quality</li><li>Prevents the model from calling on older art styles</li><li>The previous LoRAs should had never been trained on artist's name</li></ul>Dataset<ul><li>Watermark manually edited out</li><li>Primarily new images with a smaller portion of older ones included</li><li>Used to be higher but ended up with 312 imgs, did go over them a bunch. Yes, still a lot but, these are pretty consistent in style</li><li>lucy/rebecca influence brought down to increase variety and avoid same-face syndrome, at least to a certain degree</li><li>On some images missing this, manually added the subtle “line” shading (around nips) that are frequently displayed in some of their art works. This was done to increase likelihood of appearing, just because... it's hot 😎😭</li></ul>Training details<ul><li>Model trained on Illustrious XL 1.0 and WAI v15</li><li>Illustrious XL 1.0 compared against training on Noob eps and Illustrious v0.1:<ul><li>Illustrious v0.1 produced the weakest results</li><li>Noob performed well but had occasional issues with color saturation and image quality</li><li>Illustrious XL 1.0 delivered great results while avoiding the problems seen in Noob</li><li>No more pony v6, it's old and f them score tags fr. v7 maybe, previews do not give much confidence</li></ul></li></ul>

gaston18-v2-illustrious-ty_lee.safetensors

Gaston18 Style (Mashima) [Illustrious]

Illustrious v2.0

Illustrious v1.0

Pony v3.0

Pony v2.0

Pony v1.0

Illustrious Style v2I used <a target="_blank" rel="ugc" href="https://civitai.com/models/827184?modelVersionId=2167369">WAI-NSFW-illustrious-SDXL v15</a> for these images, feel free to try other checkpoints! Trained on IIllustrious XL 1.0 which is what the current WAI v15 is partly based on. Certain checkpoint or LoRA(s) may heavily influence the style.Generation details, prompts, and model info are listed with most images. CLIP IS always 2, displayed is bugged. CFG Scale 6 (5-7), 20-30 steps, and Euler a/DPM 2M as the sampler, Hires Remacri: Feel free to change!

Files with SHA256: a3f2ed97a58df8214d0839927c152dbfe369280848b3ccd081b533e6be8ef61b