[Checkpoint] Ether Real Mix

模型介绍：
🔸🌤️ Hello, this is Ether Real Mix 🌤️🔸
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! ✨

* Ether Real Mix - Version 4
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 Terry9993's 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)?
SD3 coming soon(tm)?

* Huggingface Archive

☄️ Generation Recommendations
* 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:
▣ Sampler: DPM++ 2M Karras
▣ Steps: 20 - 28
▣ CFG: 7 - 11
▣ Clip Skip: 1
▣ ENSD: 31337

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:
▣ "sketch, drawing, anime, cartoon, illustration, NSFW, (worst quality, low quality:1.3)"

* Flat Anime/Cartoon:
▣ "realistic, photorealistic, photograph, 3D, NSFW, (worst quality, low quality:1.3)"

Sample images were further upscaled using AUTO1111 Hi-res fix. Please utilize it if you wish to achieve similar results.
▣ Latent (nearest-exact)
▣ Upscale by: 2
▣ Denoising strength: 0.55

Are your results changing too much after Hi-res fix? Please try 4x-foolhardy-Remarci.

(Also recommended for simple style illustrations, anime, cartoon, etc.)

▣ 4x-foolhardy-Remarci
▣ Upscale by: 2
▣ Denoising strength: 0.45 - 0.55

Recommended VAE:
Waifu Diffusion VAE kl-f8-anime2

Strongly recommended as the model was tuned with this specific VAE's colors and saturation in mind.

🖊️ Prompting Advice
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.

▣ Generating unprompted girls/humans? Try removing "masterpiece, best quality, ....."

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:
▣ Prompt in the style of anime/booru-style models, "masterpiece, best quality, ..... "
▣ ▣ More anime style images, stronger bias towards females/humans
▣ Prompt in the style of photography/non-anime models, "professional photograph of ..... "
▣ ▣ More realistic images, higher style variety, (probably) less bias

📙 Model Merging Info - v4
I have no training or real knowledge of machine learning. I merged the model using the information provided on WarriorMama777's AOM model page as well as the ones in these Rentrys. 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 SuperMerger extension for AUTO1111 by hako-mikan.

Please refer to the 'About This Version' for more detailed information.

🧡 Credits
Thank you to WarriorMama777 for providing AOM 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 SuperMerger extension allowing a quicker workflow for merging.
Thank you to those that have contributed to these MBW Rentrys.
Thank you to gsdf, rqdwdw for Counterfeit.
Thank you to RunDiffusion for RunDiffusion FX Photorealistic.
Thank you to DynamicWang for AWPortrait and AWPainting.
Thank you to KondooAI for Juggernaut.
Thank you to Xerxemi and Xynon for SD-Silicon.
Thank you to Lykon for Dreamshaper 8.
Thank you to Terry9993 for NijiExpressive v2 and Niji v3 Fantastic.
Thank you to Ashen-sensored for Silicon Landscape Isolation.
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.

~~~~~~~~~~~~~~~~~~~~~~~o(*￣︶￣*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数：
Prompt: masterpiece, best quality, dynamic fantasy portrait illustration of a witch girl floating in the sky, scarf, witch hat, voluminous puffy clothing, floating hair, lace trim, warm soft lighting, looking at viewer, smug, smirk, sunset, cloudy sky, (light particles:1.1), volumetric lighting, sparkle, sparkling, star, glittering, expressive, iridescent

Negative prompt: NSFW, (worst quality, low quality:1.3)

Steps: 21, CFG scale: 12, Sampler: DPM++ 2M, Seed: 918830258, VAE: kl-f8-anime2.vae.pt, Size: 512x896, Model: EtherReal-v4-LORAmerge-final.fp16, Version: v1.9.3, Model hash: 081c0d5fa1, Hires upscale: 2, Schedule type: Karras, Hires upscaler: Latent (nearest-exact), Denoising strength: 0.55