This is just a fun little experiment/mix by simple merging 25% of: LuminEdge V0.5, Arcazelium V0.6, Ecliptica-ArcaneX V1.0 - Refined, and Finiticore Fusion V3.0, which they are my favorite versions of those models. This model was to create a mix out of curiosity, I uploaded it cause why not :-P.Have fun experimenting if you want. Beta Recommended Settings: Steps: 25+ CFG: 4+ Sampler: (Haven't tested past Euler) Scheduler: (Haven't tasted past Normal) Quality tags: Positive Prompt) score_9, score_8_up, score_7_up, (masterpiece), best quality Negative Prompt) score_6, score_5_up, score_4_up, worst quality, old, early, low quality, lowres, signature, artist name, signature, username, logo, bad hands, mutated hands, bokeh, simple background, jpeg artifacts, young Note: You will need the scores as the style is mainly realism pony with 40% pony clip possibly more I didn't keep track. Alpha Recommended Settings: Steps: 25+ CFG: 4+ Sampler: (Haven't tested past Euler) Scheduler: Haven't test past Normal) Quality Tags: Positive Prompt) score_9, score_8_up, score_7_up, (masterpiece), best quality Negative Prompt) score_6, score_5_up, score_4_up, worst quality, old, early, low quality, lowres, signature, artist name, signature, username, logo, bad hands, mutated hands, bokeh, simple background Note: Most of the mix it seems is SDXL-Pony with 25% illu/noobai. So you'll have to use scores.

Orcaventhis

A much better version, this was with block merges and simple merges featuring the same models. Also includes block merges of other models that aren't mine. Alpha version was to see the potential, if any, that it had. This also has more of a realistic style than anything else. File is embedded with the workflow to create this, like always. Note: Most (90%) of the style is a mix of GreatHunt's first two models. Plus file is embedded with the workflow to create the merge model

Files with SHA256: 61e14aa3204657ce2edefe0c9fcd870ebd9412232b0a890e30e64e27b5207445