★🧪 Entering: the Ether Lab 🧪★
Ether Lab: A collection of experimental and WIP models that still feel fun to use or don't belong in a specific 'Ether' release. Please note that these models are unfinished, may behave strangely and likely exhibit other potential quirks.
✨ Please Share Your Cool Creations Below! ✨
* Update: One-WIP-v2
This update looks to further improve the model's aesthetic quality and appeal while retaining its focus on flexibility. I pushed plenty of weights within the more aesthetically tuned models as well as adding Coco Style Illustrious, which sadly seems to have been removed.
This is probably (hopefully) the last WIP iteration as I plan to go into an even heavier aesthetic focused merge, likely starting with a full (?) release of Ether One Illustrious.
Life has picked back up again so apologies in advance if updates are sporadic and far in between. As always thank you so much for your continued love and support.
I recommend describing the style (oil painting, messy sketch, anime screenshot etc), adding artist tags and being more descriptive with your prompts. Feel free to mix natural language and booru tags.
Looking forward to your feedback!
☄️ 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.
Most sample images were generated with:
HxW: 832 x 1216
Sampler: Eular A
Schedule Type: SGM Uniform
Steps: 28 - 36
CFG: 6 - 9
Clip Skip: 1
Quality Tags:
masterpiece, best quality, amazing quality, very aesthetic, absurdres, newest
Negative Prompt Example:
worst quality, low quality, bad quality, bad anatomy, lowres, watermark, signature
I recommend this 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.
Sample images were further upscaled using Hi-res fix. Please utilize it if you wish to achieve similar results.
Latent (nearest-exact)
Upscale by: 1.5
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: 1.5
Denoising strength: 0.45 - 0.6
🚧 Information page is WIP, thanks for your patience. 🚧
📘 Model Merging Info - One-WIP-v1
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 a mix of the SuperMerger extension for reForge by hako-mikan and silveroxides fork of sd-webui-untitledmerger.
Please refer to the 'About This Version' for more detailed information.
💙 Acknowledgements
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 and this very detailed MBW Figma.
Thank you to AngelBottomless and OnomaAI for Illustrious XL 2.0.
Thank you to gsdf, rqdwdw for Obsession (Illustrious-XL) v3.1.
Thank you to Minthybasis for Rouwei v0.8.0 epsilon.
Thank you to asyncforai414 for Async's MIX ILLUSTRIOUS IL V4.0.
Thank you to EWIGKEIT0 for DREAMBOX v3.5 and NTDMIX v1.5.
Thank you to AUTO1111 for creating the web UI.
Thank you to Panchovix for reForge.
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 Stable Diffusion and generative art community for continuing to openly share and create.
🚧 Information page is WIP, thanks for your patience. 🚧