AndHor/JustAnotherFurryMerge

JustAnotherFurryMerge

JustAnotherFurryMerge (JAFM) is my personal merge I made using ComfyUI. It first started as experiment with merging random furry models I have. Turns out it gens very well so I decided to share it with anybody out here. It is intended to generate furry anthros. It can do males and possibly anime as well, however I'm not into that. Using mostly e6 tags for my gens.

This merge is eps prediction. It has a piece of v-pred in it, but since basemodel is eps the output model is eps as well.

It can do both sfw and nsfw just fine, sample images here are sfw only. All have workflow saved in meta, should be reproducible.

Recommended settings:

• Steps: 10 - 30 (16+ for zoom-in better quality)
• CFG: 2 - 6 (6+ if you like oversaturation)
• Sampler: Euler a, dpm++ 2m sde, ddpm, seeds 2
• Scheduler: beta > karras > normal
• Prompting: Using quality tags seems to improve quality in some way. Probably does not matter if you put them on beggining or end of prompt. Use whatever you prefer.
• Resolutions: Anything between 0.9 and 2.4 MPx is okay. Unless too vide or tall. Or try this :)

I personally gen with Euler a@beta on 30 steps@3 cfg at 1080p + 2-pass img2img upscaling to 4k with dpm++ 3m sde gpu@karras on 16 steps only and 0.16 denoise, same cfg (v4.0 showcase workflow). Takes a while but deffo worth the wait. I don't guarantee it to work under 32GB ram/12 vram, anyway nothing stops you from trying it :P

Running showcase workflows:

First make sure you have latest version of ComfyUI with Manager with and all custom nodes installed. Copy pasting again from Civitai should not throw any missing nodes. Don't forget to check suggested resources for detectors I used. They need to be placed in "models/ultralytics/segm/" on the same drive your ComfyUI is saved. Having "ultralytics: models/ultralytics/" in "extra_model_paths.yaml" file does not work and probably never will.

If workflows from older versions doesn't run with correct model/vae selected, try using from never ones, copy prompt, seed and choose model. This might happen as ComfyUI with customnodes getting frequent updates. I also genned v2.0 showcases on high resolutions. Anyway I changed workflow on some showcases in v3.0 and v4.0 are all 4k upscaled.

Version history:

• 1.0 & 1.0unstyled - First release

  First version (4 models + 2 loras)
  
  Base = Alien Furry Mix IL * 0.32 +
  Nova Furry v7B * 0.32 +
  SteinCustom v7 * 0.32 +
  Manticore V-pred v1.0 * 0.04
  
  JAFM v1.0 = Base + Great Lighting * 1 +
  IndigoToonMix Style IL * 0.5 (not in unstyled versions)
  

• 2.0 & 2.0unstyled - Hires rawgens quality improvement

  Attempt to improve quality of hi-res rawgens (6 models)
  
  JAFM v2.0 = v1.0 * 0.55 +
  Illustrious XL 2.0 * 0.10 +
  Yiff in Hell v2.0 * 0.10 +
  MixIllustriousMerge1 6th * 0.10 +
  Tiger Milk v1 * 0.10 +
  Auronami v1 * 0.05
  

• 3.0 & 3.0unstyled - Small knowledge enchancement

  Small knowledge enhancement (7 models + 7 loras) + my own Better Colors Vae Monumental
  
  Base = v2.0 * 0.6 +
  Arctenox's Personal v0.5C * 0.06 +
  NovaFurry IL v14.0 * 0.08 +
  YobboIllustrious v2 * 0.06 +
  Augenstein v5B * 0.06 +
  Ultra4illustrious v1.0 * 0.07 +
  Cinema l'usine Lumière * 0.07
  
  JAFM v3.0 = Base + PornMaster - rule 34 prompts v2 * 0.1 +
  Add Micro Details v5.0 IL * 0.05 +
  Add Micro Details v3.0 NAI * 0.05 +
  NoobAI-XL Detailer EPS1.1 v1.0 * 0.6 +
  Detail Tweaker XL v1.0 * 0.5 +
  Eyes High Definition v1 * 0.5 +
  Hands XL v5.5 * 0.75
  
  Rule 34 & Hands loras permission to merge granted - Big thanks!
  

• 4.0 & 4.0unstyled - Enhancing the style

  Enhancing the style with 4 premixes (20 models) (possibly last version for now)
  
  Premix A = Uwumerge v5.0 * 0.30 +
  Plant Milk Flax * 0.25 +
  Semi-Realistic Furry v1.0 * 0.30 +
  Furstein V2A1 * 0.10 +
  Furreal Dreams v1.0 * 0.05
  
  Premix B = Chenkin Noob v0.5 * 0.35 +
  Wild Safari v2.0 * 0.20 +
  Banana Splitz XXL v1.1 * 0.15 +
  Reed Furry Mix v2.3nai * 0.20 +
  Yiff in Hell Vantablack * 0.10
  
  Premix C = Hyphoria v0.02 * 0.30 +
  Nova Retro XL v3.0 * 0.20 +
  DeepNavy v3 * 0.25 +
  IcedCoffeeIL v4.0 * 0.15 +
  Plant Milk Walnut * 0.10
  
  Premix D = Arctenox's Personal v0.5B * 0.25 +
  Nova Furry XL v18.0 A * 0.25 +
  Arctenox's Personal v1.0B * 0.25 +
  Nova Furry XL v18.0 B * 0.10 +
  Scruff silver blend v1.3 * 0.15
  
  JAFM v4.0 = v3.0 * 0.60 +
  Premix A * 0.12 +
  Premix B * 0.05 +
  Premix C * 0.10 +
  Premix D * 0.13
  

Try int8 convrot quants, saves 2GB memory without much changing output! (That surprised me actually)

Credits

Thank you all for creating such nice models! Without you, my own would not be possible:

Checkpoints: Agri_Yob, AI_nstein, Anomalous_Anonymous, BlessedBooba, Chenkin, Crody, DarkFawkes, ecaj, EJtech, GreyFerret, iDeNoh, jhonny09, kftiger, LatteLeopard, Mol_Keun, multilazor, Ocean3, Old_Arctenox, ONOMAAI, ReedVaughn, RodAndRog, SXCD2719, thebiologist92137
Loras: Anzatiridonia, EauDeNoire, iamddtla, Meower2024, ntc, UnholyDesiresStudio, Volnovik, w4r10ck
Vae: First 2 releases stabilityai, ongoing myself

This respository also has beta versions of the merge. Using betas (and int8 convrot) for merging is forbidden. For rules on full versions please visit CivitAi.

Post your gens, share if you like it! And don't forget, no reprints please.

And as always

>> Just keep generating
>> Don't ever try to stop
>> Share your best works
>> Seek new motivation