Model trained with 100 images ( 50 + 50 fliped ) using the Seek Art Mega model as a base model before training and later used to "fix" the damage caused by dreamboothing through multiple merges: <a target="_blank" rel="ugc" href="https://civarchive.com/models/1315/seekart-mega">https://civarchive.com/models/1315/seekart-mega</a> Workflow and more samples with according tested artists: <a target="_blank" rel="ugc" href="https://www.artstation.com/artwork/2qPoAx">https://www.artstation.com/artwork/2qPoAx</a><ul><li>Trained with some empty backgrounds with different solid colors to easily produce removable backgrounds;</li><li>Stacked training with stacked merges using the base model to avoid loss of other artstyles;</li><li>Most of the trained images contains tattoos, jewelry, stamps, and detailed equipment of scifi fantasy fashion;</li><li>Trained with micro crops of large compositions and according original compositions to easyly insert the characters within scenarios;</li></ul>

CCDHT - Concept Character Designer - Oriental

The raw file already merged with <a target="_blank" rel="ugc" href="https://civitai.com/models/4118/spybgs-toolkit-for-digital-artists">SPYBG Toolkit</a> ( just for ease of use if you don't know how to merge models ), then, after that, another dreambooth training session to restore the lost artstyle. That model is highly biased towards 768x768 resolutions, anything below that resolution scale will only create ugly outputs, and strongly biased towards realistic / photografic characters ( loosing artstyles ), however:<ul><li>Fucking good to generate character reference sheets and character turnarounds ( without even asking while prompting );</li><li>High quality hands as long as you don't prompt to pose them ( just let the model decide how the hands should be placed );</li><li>Tiling / motif probability ir nearly zero ( duplicated features );</li><li>Reduced amount of intricate details ( unless you prompt for );</li><li>Largely resilient to alternating keywords without loosing coherence;</li><li>Trained tokens inherited from original model;</li><li>Since most of the artsyles bias towards the same look, it's easy to blend lots of artists without causing noisy features or incoherence;</li><li>Tends to "obey" your prompts without a huge CFG scale;</li></ul>

Files with SHA256: 4fea17edb69b6098e5701506c53b29d14e439bf4f584217988410694382fe539