โ ๐ Hello, this is Ether Moonlight Mix ๐ โ
A fun little anime style model designed to replicate one of my favorite mediums - oil paintings!
Please keep in mind that this model has biases in generating females and humans.
โจ Please Share Your Cool Creations Below! โจ
* Beta 3
Updated Ether One and added a pinch of Ether Blu 7 to increase vibrancy, as well as making illustrations more expressive. Lots of weight changes designed to further refine the painterly style. Tuned faces to be a little more semi-realistic, though still very much anime. Finished off with some LORA's to compliment the overall style.
Feedback will be appreciated!
I recommend utilizing words that describe mediums, such as oil painting, watercolor, gouache, which can further enhance the painterly effect. Additionally, evoking painters can do similar things. Try great painters and illustrators such as John Singer Sargent, Dean Cornwell and Anders Zorn.
Artist and style resources are a great way to both tune the style of your outputs and learn more about art.
* Huggingface Archive
โ๏ธ Generation Recommendations
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), signature"
I recommend this simple negative prompt if you wish to achieve a similar style to the sample images.
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:
The VAE is baked.
Waifu Diffusion VAE kl-f8-anime2
Strongly recommended as the model was tuned with this specific VAE's colors and saturation in mind.
๐๏ธ Prompting Advice
As this is an anime style model I tuned it with booru style tag prompting, 1girl, long hair, jacket, etc. Of course you're free to prompt however you like. Mixing tags with prose can lead to fun results.
Utilizing words that describe mediums, such as oil painting, watercolor, gouache, can further enhance the painterly effect. Additionally, evoking painters can do similar things. Try great painters and illustrators such as John Singer Sargent, Dean Cornwell and Anders Zorn.
Artist and style resources are a great way to both tune the style your outputs and learn more about art.
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
๐ Model Merging Info - Beta 3
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.
๐ 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 creating Counterfeit.
Thank you to Xerxemi and Xynon for SD-Silicon.
Thank you to Mushroom for Oil painting.
Thank you to Yang for Watercolor Oil Painting.
Thank you to Someone97421 for Oil painting(oil brush stroke) - ๆฒน็ป็ฌ่งฆ and Fechin oil painting - ่ดนๆฌฃๆฒน็ป
Thank you to Aotsuyu for Darkv7
Thank you to OliviaOliveira for Olivian Opulance.
Thank you to those at the LMFResearch Society for their archive of SD LORA's and checkpoints.
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.
Description
* Beta 1
An initial test release that, while not perfect, can still produce fun results.
Hey everyone. I mixed this model as a little side project. I wanted an anime, booru tag model that would output a style with thick brushwork reminescent of oil paintings.
I recommend utilizing words that describe mediums, such as oil painting, watercolor, gouache, which can further enhance the painterly effect. Additionally, evoking painters can do similar things. Try great painters and illustrators such as John Singer Sargent, Dean Cornwell and Anders Zorn.
Artist and style resources are a great way to both tune the style your outputs and learn more about art.
Still a lot to work on and improve, but I hope you have fun with what it can do!
๐ Model Merging Info - Beta 1
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.
โผ Merging
MODEL A: Counterfeit-V3.0_fp16.safetensors
MODEL B: Silicon29-dark.safetensors
Weight Sum merge - โ Don't Use MBW
Alpha: 0.5
Model: Counterfeit3-Silicon29DarkSum-05
โผ
MODEL A: Counterfeit3-Silicon29DarkSum-05
MODEL B: oilPainting_oilPaintingV10-pruned
MODEL C: watercolorOilPainting_v10-pruned-clipfix
Triple Sum merge - โ๏ธ Use MBW
Alpha: 1
0,0.25,0.25,0.25,0.25,0.25,0,0,0,0,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.15,0.15,0.25,0.25,0.25,0.25,0.25,0.25,0.25
Beta: 1
0,0.75,0.75,0.75,0.75,0.5,0.25,0.1,0.1,0.1,0.25,0.25,0.25,0.25,0.5,0.5,0.5,0.35,0.35,0.5,1,1,1,1,1,1
Model: Counterfeit3-Silicon29DarkSum-05-Oilpainting-WatercolorOil
โผ LORA Mix:
bichu-v0612 - 0.3
fechin - 0.4
darkv7 - 0.2
Final Model: Ether Moonlight Mix
Models Used:
*Counterfeit-V3.0_fp16.safetensors
CBFBA64E662370F59D4AA2AA69BF16749FCE93846CCCE20506AEE5DF01169859
*Silicon29-dark.safetensors
de329f31ad222a0d765897aa38796811276691288b104099523649bdf1ef9708
*oilPainting_oilPaintingV10.safetensors
9DA0C2B94D26D769E855C3739D8D69A25BEEEE94A9F8A2573D30CD852DDCA438
*watercolorOilPainting_v10.safetensors
0CBEA7D5041FC10F273D148A89D56F297D5919059F864040F17F3ABF709B410BFAQ
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.