sexual situations

physical violence

disturbing

male nudity

extremist

hanging

hate symbols

nazi party

partial nudity

white supremacy

adult toys

graphic male nudity

illustrated explicit nudity

nudity

self injury

graphic female nudity

peeing

exposed female nipple

suggestive

weapon violence

female swimwear or underwear

male swimwear or underwear

emaciated bodies

sexual activity

sexual intent

revealing clothes

graphic violence or gore

corpses

wide hips

convenient censoring

oral invitation

blowjob

female nudity

undressed

male underwear

female swimwear

genitals

female underwear

thick thighs

breasts out

strapless leotard

vore

bestiality

breast out

one breast out

huge breasts

gigantic breasts

huge butt

covered nipples

hair over breasts

no panties

sitting on face

anal

dildo riding

downblouse

oral

porn

futanari

hentai

nude

lingerie

sexy

nsfw

bukkake

cumshot

cumdrip

cum in pussy

cum on face

cum on hair

cum on body

cum on tongue

cum on hands

cum in mouth

cum on pussy

eat_cum

fellatio

implied fellatio

after fellatio

triple fellatio

autofellatio

pov fellatio

fucked silly

presenting ass

presenting pussy

spreading legs

spread legs

swimwear

<h1 id="v3.0:-qjn0encti">V3.0:</h1>2 months of work, and I think 3.0 is finally ready for showtime:A completely new redesign built using different ideas, using a primarily human only model and applying 'patches' as it were for doing what is essentially a furry conversion yielding much improved detail.Tips below still apply, and some tags might spawn humans where my tip would be to try a different word with the same meaning, say a booru/e621 tag of said subject.Or try the furryfication slider lora. But that is best used with solo subjects.Of note: the bigger version should help with fur and background detail and such, if you can it is recommended to use it instead of the smaller float16 converted one.Many excel sheets, Visio documents, 6 gigabytes of test images, possible 300 or more GB of test models made over the span of the 2 to 3 months.And yes it is still tagged as a Pony model even if I am pretty sure it has become so diluted at this point it might not respond nowhere near the strength it used to. But hey, try it anyway.If you end up getting too many humans, check the length of your negative prompt or other non-booru tags that might give problems if the underlying base model were to respond too much to it. Recommending to keep thing short and simple.Or try v2.2 and see if it works better for you if you are not chasing the 'best quality' as I am :P<hr />Models used:<ul><li>Base: <a target="_blank" rel="ugc" href="https://civitai.com/models/989068?modelVersionId=1108043">Brehmsstrahlungmix 1.0</a></li><li>Furry additives: <a target="_blank" rel="ugc" href="https://civitai.com/models/310964?modelVersionId=558568">Furry Enhancer 6.1</a> + <a target="_blank" rel="ugc" href="https://civitai.com/models/200230?modelVersionId=991039">OneForAll Pony Reality</a> + <a target="_blank" rel="ugc" href="https://civitai.com/models/797769?modelVersionId=1059787">Lucid Citrus Pony Kiwi</a> + Porn Master BDSM (a rule34 dataset, got removed from CivitAI) + <a target="_blank" rel="ugc" href="https://civitai.com/models/1131256?modelVersionId=1284061">Quintillus </a>+ <a target="_blank" rel="ugc" href="https://civitai.com/models/153568?modelVersionId=1308507">RealDream 14</a></li><li>Negatives: <a target="_blank" rel="ugc" href="https://civitai.com/models/617652?modelVersionId=1030169">Natural Vision 2.7</a></li></ul><hr />Results so far:<ul><li>Better lighting</li><li>Background detail and items, say at an office a whole computer setup appears</li><li>Better usability of sentences and tag based prompts, see our standard dragon lady who is no longer brown and has the correct expression.</li><li>Plus whatever else. Quirks are still being worked out to figure out what to finetune exactly.</li><li>Shoutout @freek22 for<a target="_blank" rel="ugc" href="https://civitai.com/models/310964?modelVersionId=558568"> Furry Enhancer 6.1</a></li><li>Don't forget to join the Furry Diffusion Discord for discussions and posts of everyone's work: --&gt; <a target="_blank" rel="ugc" href="https://discord.gg/furrydiffusion">https://discord.gg/furrydiffusion</a> (I'm mostly active there)</li></ul><hr /><h3 id="v2.x:-sdxl-pony-compatible-co5h8oevy">V2.X: SDXL, Pony compatible</h3>Having problems with rampant humans everywhere? Try adding source_furry and/or 'furry art'.<h3 id="tip:-29kd58ndu">Tip:</h3>As some have asked for help for a point to start your prompt from, something like:<pre><code>source_furry, source_9, score_8_up, score_7_up, furry art,
Rest of your prompt here</code></pre>Optional <code>photorealism, realistic,</code> and for negative;<pre><code>low quality, bad composition, blurry, bad anatomy, bad proportions,ugly, lacking, mediocre, mutation, conjoined, fused, clone, mirrored, disjointed, incomplete, watermark, signature, artist name, camera, extra anus, skin, human feet, ((human, humans)), score_4,</code></pre>Should be a good starting point. I have noticed over the last long while, depending on the subject you're trying to do having too many quality tags can be detrimental so a KISS approach, (keep it simple, **) would be good.<h3 id="older-notes:-04bifb6d9">Older notes:</h3>New v1.9 after 8 months!:<ol><li>Fluffyrock tags</li><li>Many hundreds of tests run in total over 5+ models</li><li>Should be an excellent allrounder.</li></ol>Now back to the old stuff:This model mix was found by 'accident' a month or two ago and has decent requests on Furry Diffusion for it to be available for download so here it is. It can do both safe and mature works.The mix is supposed to handle traditional art and classic 'toony' furry styles, but as the base models used for mixes are seemingly moving away from that it is a good idea to try all versions if you cannot get something specific to work right.As such: 1.2 is a good starting point. 1.4 could/will result in different poses and scenes.But with that said, the fault is obvious and I am working almost every day for more, better tuning, finding the faults or strength of the model.For instance:<ul><li>Extremely high background detail, even without LoRa's. Especially when using hires fix can get very fancy things built into the background of whatever scene you, the director is trying to create.</li><li>Newer models might have issues with eyes, as the artist data gets a bit 'messy'. Thus, it is recommended to explicitly mention what color eyes you want, and pupils if applicable. Adding or removing artist tags can also help a ton in this case.</li><li>Extremely niche Kinks/fetish works might need older model versions like 1.2, this is a known thing so I'm working on it.</li></ul>Otherwise, new models are at the mercy of others since this is a tuned mix of other tuned mixes for maximum detail and universability, or that is the idea.

v3.1:Iterative release, uses some new tech and an updated base model plus the new Furry Enhancer 8.2 by ye olde freek22 to get things a bit higher, sharper, more attentive.Improved fur and scale uniformity, teeth, background basics, basically it should all be a bit better or at the worst, a little different.

v3.1

v3.0

V3.0 Stage 0:A new base model built using different ideas, using a primarily human only model and applying 'patches' as it were for doing what is essentially a furry conversion.So this is the base model as it were, stage 0 as one would say in a way. All the other mixes using v2 failed or did not offer enough upgrades and some people also commented how similar these models are getting so that's why I did it like this.Fine tuning has not happened yet, I am still working on my approach and what things need fixing. Tips below still apply, and some tags might spawn humans where my tip would be to try a different word with the same meaning, say a booru/e621 tag of said subject.Also of note, the bigger version DOES help with fur detail and such, if you can it is recommended to use it instead of the smaller float16 converted one.Results so far:<ul><li>Better lighting</li><li>Background detail and items, say at an office a whole computer setup appears</li><li>Better usability of sentences and tag based prompts, see our standard dragon lady who is no longer brown and has the correct expression.</li><li>Plus whatever else. Quirks are still being worked out to figure out what to finetune exactly.</li></ul>Work in progress!<ul><li>Shoutout @freek22 for<a target="_blank" rel="ugc" href="https://civitai.com/models/310964?modelVersionId=558568"> Furry Enhancer 6.1</a> and Ratatoskr 8 :P</li><li>Don't forget to join the Furry Diffusion Discord for discussions and posts of everyone's work: --&gt; <a target="_blank" rel="ugc" href="https://discord.gg/furrydiffusion">https://discord.gg/furrydiffusion</a> (I'm mostly active there)</li></ul>

v3.0-Stage-0

v2.2!Better lighting, fur, scales, less glitches and not completely different with the work you'd already have. Plus my first two parter mix using multiple models to compliment each other!

v2.2

v2.1

XL version! Took a while to find a combination that seemed beneficial.Pony-compatible base and mixes, detailed backgrounds can get a bit dark at times.

v2.0

Holy heck a new version. Yeah. Been a while. Anyway, finally something worth sharing!1.9 is a triple mix of three models, (I'm pretty sure the recipe is saved in the model), all the previous attempts got some results over the last few months but never quite that 'wow' that the first one did. Even if I kept that one to myself for a bit too long, and by the time I did share it, things got better than it. Oh well. Click show more!I noticed most SD 1.5 non vpred models were stopping so ironically here is my mix in what I can call the final v1. Combining the models I used the most, tested the most, using a triple mix to get what ended up being the best of what all did. In the main Excel sheet, the base won 6 times, the mix won 14 times in a range of prompts and scenarios.Not vpred, no configs required. Just download and go. Fluffusion VAE baked in. Leave your settings (on say, a111) to default and it should grab it.

v1.9

Right, 1.4. This one took a couple weeks with around 12 to 20 hours of combination and weight tuning. Excel sheet to keep track of which one was better and everything.The goal remains the same, just now with better eyes, background detail, and not completely deleting portions or tags that worked in previous versions.Enjoy and see what you can crank out with this one!

v1.4-Tuned

A version made within a very short amount of time with minimal testing because the other model I was using got updated a week later, thus I'm working on 1.4 instead, which is requiring a lot more work due to changes on the base models.For more information, see the v1.0 release notes!

v1.3-Test

This version has a bit of a 'sweetspot' in mind, but that could be because I used this one for quite a long while before 1.3 and 1.4 were made in rapid succession.

v1.2

First version, resulted in massively boosted background detail.

v1.0

Hayden's YiffMix Realism

10327-493752147 (posted).png