<h1 id="hello-this-is-ether-real-mix-l0hg1u80l">🔸🌤️ Hello, this is Ether Real Mix 🌤️🔸</h1>Ether Real Mix is a stylized realism model focused on flexibility. It's capable of producing a variety of subjects in multitudes of styles. It still has some shortcomings inherit to anime models such as light biases in generating females and humans.This model is intended to act as a 'blank canvas'. Add your favorite LORA's and embeddings to customize it to your liking!✨ Please Share Your Cool Creations Below! ✨<h3 id="*-ether-real-mix-version-4-5w9405zf6">* Ether Real Mix - Version 4</h3>Ether Real Mix 4 looks to further refine the realistic/semi-realistic aesthetic through LORA block merging, updating ingredients, and (hopefully) improving anatomy and hands within the constraints of a realistic SD 1.5 model.I shuffled out a lot of models that felt redundant or just no longer fit the goals of Ether Real. The aesthetic is tuned towards a Niji inspired feel thanks to <a target="_blank" rel="ugc" href="https://civarchive.com/user/Terry9993">Terry9993's</a> Niji styled LORA's. Overall, I'm looking to strike a balance between anime-esque realism while also not having outputs that feel too derivative of Nijijourney.VAE is baked in.More detailed list of changes coming soon(tm).SDXL/Pony model coming soon(tm)?<s>SD3 coming soon(tm)?</s><h3 id="-ye61tx121"></h3><h3 id="*-huggingface-archive-0x1czxi7t"><a target="_blank" rel="ugc" href="https://huggingface.co/gamerdan69/EtherMix">* Huggingface Archive</a></h3><h3 id="-fpnst4s8t"></h3><h2 id="generation-recommendations-vm6814rdd">☄️ Generation Recommendations</h2>* All preview images were generated with no LORA's, embeddings, extensions or img to img/inpainting. Just pure text to image with hi-res fix upscaling.Use whatever sampler, steps, CFG you prefer.Most sample images were generated with:<ul><li>Sampler: DPM++ 2M Karras</li><li>Steps: 20 - 28</li><li>CFG: 7 - 11</li><li>Clip Skip: 1</li><li>ENSD: 31337</li></ul>Higher CFG can lead to more vibrant results.Negative Prompt Example:"NSFW, (worst quality, low quality:1.3)"I recommend this simple negative prompt if you wish to achieve a similar style to the preview images. You can further alter positive and negative prompts based on the type of style you'd like to achieve.Style focused negative prompt examples:* Photorealism feel:<ul><li>"sketch, drawing, anime, cartoon, illustration, NSFW, (worst quality, low quality:1.3)"</li></ul>* Flat Anime/Cartoon:<ul><li>"realistic, photorealistic, photograph, 3D, NSFW, (worst quality, low quality:1.3)"</li></ul><h3 id="-d5kf17rys"></h3>Sample images were further upscaled using AUTO1111 Hi-res fix. Please utilize it if you wish to achieve similar results.<ul><li>Latent (nearest-exact)</li><li>Upscale by: 2</li><li>Denoising strength: 0.55</li></ul>Are your results changing too much after Hi-res fix? Please try <a target="_blank" rel="ugc" href="https://openmodeldb.info/models/4x-Remacri">4x-foolhardy-Remarci.</a>(Also recommended for simple style illustrations, anime, cartoon, etc.)<ul><li>4x-foolhardy-Remarci</li><li>Upscale by: 2</li><li>Denoising strength: 0.45 - 0.55</li></ul><h3 id="recommended-vae:-bzcb1hvkr">Recommended VAE:</h3><a target="_blank" rel="ugc" href="https://huggingface.co/hakurei/waifu-diffusion-v1-4/blob/main/vae/kl-f8-anime2.ckpt">Waifu Diffusion VAE kl-f8-anime2</a>Strongly recommended as the model was tuned with this specific VAE's colors and saturation in mind.<h3 id="prompting-advice-9aw6j0ton">🖊️ Prompting Advice</h3>Prompt however you like. The model was merged with both Danbooru-style tag prompting as well as natural language in mind. You're free to mix and match prompting styles, the model should understand it.<ul><li>Generating unprompted girls/humans? Try removing "masterpiece, best quality, ....."</li></ul>If you look at the prompts included with the sample images they demonstrate that even with a very loose and vague style of prompting you can achieve fun, high quality images.Try mixing natural language and booru-styled tags. Experiment and give the AI some freedom to interpret your ideas!General guidelines:<ul><li>Prompt in the style of anime/booru-style models, "masterpiece, best quality, ..... "<ul><li>More anime style images, stronger bias towards females/humans</li></ul></li><li>Prompt in the style of photography/non-anime models, "professional photograph of ..... "<ul><li>More realistic images, higher style variety, (probably) less bias</li></ul></li></ul><h3 id="-qhy2n0wzw"></h3><h2 id="model-merging-info-v4-ick57ja67">📙 Model Merging Info - v4</h2>I have no training or real knowledge of machine learning. I merged the model using the information provided on <a target="_blank" rel="ugc" href="https://huggingface.co/WarriorMama777/OrangeMixs">WarriorMama777's AOM model page</a> as well as the ones in <a target="_blank" rel="ugc" href="https://rentry.org/BlockMergeExplained">these</a> <a target="_blank" rel="ugc" href="https://rentry.org/Merge_Block_Weight_-china-_v1_Beta">Rentrys.</a> I very arbitrarily mixed the UBM weights of each model, modifying the values with trial and error until I was visually satisfied with it.I utilized the <a target="_blank" rel="ugc" href="https://github.com/hako-mikan/sd-webui-supermerger">SuperMerger</a> extension for AUTO1111 by hako-mikan.<h3 id="please-refer-to-the-'about-this-version'-for-more-detailed-information.-m8s8wzejv">Please refer to the 'About This Version' for more detailed information.</h3><h3 id="-ucd05uwzb"></h3><h2 id="credits-17sxip77z">🧡 Credits</h2>Thank you to WarriorMama777 for providing <a target="_blank" rel="ugc" href="https://civarchive.com/models/9942/abyssorangemix3-aom3">AOM</a> and other various mixes as well as detailing your workflow which inspired me to try mixing my own models.Thank you to hako-mikan for creating the <a target="_blank" rel="ugc" href="https://github.com/hako-mikan/sd-webui-supermerger">SuperMerger</a> extension allowing a quicker workflow for merging.Thank you to those that have contributed to these <a target="_blank" rel="ugc" href="https://rentry.org/BlockMergeExplained">MBW</a> <a target="_blank" rel="ugc" href="https://rentry.org/Merge_Block_Weight_-china-_v1_Beta">Rentrys.</a>Thank you to gsdf, rqdwdw for <a target="_blank" rel="ugc" href="https://civarchive.com/models/4468/counterfeit-v25">Counterfeit.</a>Thank you to RunDiffusion for <a target="_blank" rel="ugc" href="https://civarchive.com/models/82972/rundiffusion-fx-photorealistic">RunDiffusion FX Photorealistic.</a>Thank you to DynamicWang for <a target="_blank" rel="ugc" href="https://civarchive.com/models/61170?modelVersionId=304593">AWPortrait</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/84476/awpainting">AWPainting.</a>Thank you to KondooAI for <a target="_blank" rel="ugc" href="https://civarchive.com/models/46422/juggernaut">Juggernaut.</a>Thank you to Xerxemi and Xynon for <a target="_blank" rel="ugc" href="https://huggingface.co/Xynon/SD-Silicon">SD-Silicon.</a>Thank you to Lykon for <a target="_blank" rel="ugc" href="https://civarchive.com/models/4384?modelVersionId=128713">Dreamshaper 8.</a>Thank you to Terry9993 for <a target="_blank" rel="ugc" href="https://civarchive.com/models/47909/nijiexpressivev2">NijiExpressive v2</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/130311/niji-cool-perfect-anime-lora">Niji v3 Fantastic.</a>Thank you to Ashen-sensored for <a target="_blank" rel="ugc" href="https://huggingface.co/ashen-sensored/lora-isolation-collection">Silicon Landscape Isolation.</a>Thank you to AUTO1111 for creating the web UI.Thank you to StabilityAI for starting everything with Stable Diffusion.Thank you to CivitAI and Huggingface for providing places where open source AI can flourish.Thank you to the entire SD community for continuing to openly share and create.

Ether Real Mix

* Ether Real Mix - Version 3Another rebuild because... I've started learning MBW merging!Since I felt I've reached some limitations with my current method of merging, I've decided to actually learn about Unet block merging. I've tuned the weights for each model according to what I'd like out of them, style, color, realistic texture and lighting, etc.Firstly, I substituted AOM3 for <a target="_blank" rel="ugc" href="https://huggingface.co/Xynon/SD-Silicon">Silicon29Dark</a>, an auto-merged anime model with noise offset. This should allow for better contrast and darker low-light images. Since AOM3 is focused on NSFW content I wanted an anime model that was more generalized.DalcefoPaint was substituted with <a target="_blank" rel="ugc" href="https://civitai.com/models/84476/awpainting">AWPainting</a> as I preferred its facial style, linework and lighting. <a target="_blank" rel="ugc" href="https://civitai.com/models/46422/juggernaut">Juggernaut</a> is then added for its realistic texture and lighting. <a target="_blank" rel="ugc" href="https://civitai.com/models/5414">Pastel-Mix</a> is added for its colors, <a target="_blank" rel="ugc" href="https://civitai.com/models/4550/ayonimix">AyoniMix</a> for its semi-realistic style and <a target="_blank" rel="ugc" href="https://huggingface.co/nuigurumi/basil_mix">Basil Mix</a> to help with anatomy.Overall, ERM3 should have better, stronger lighting and texture, with maybe a touch more bias (sorry). Should hopefully still be quite flexible.Hoping to iron out quirks and improve Ether Real as I continue to learn MBW merging.

Files with SHA256: 6e8c828c9f32874edb9747eaecce44c161e32c84db3efc5d8252b12362314688