For business inquires, commercial licensing, consultation and large scale image captioning for datasets contact me under [email protected], for custom LoRAs visit my ko-fi page.

𝓒𝓱𝓮𝓬𝓴𝓹𝓸𝓲𝓷𝓽 𝓣𝔂𝓹𝓮: 𝓓𝓮𝓯𝓲𝓷𝓲𝓽𝓲𝓿𝓮 𝓢𝓓𝓧𝓛 𝓜𝓞𝓓𝓔𝓛

Follow me on Twitter: @YamerOfficial Discord: yamer_ai

If you have enjoyed this checkpoint:

It is time to create an all Yamer-style model, and what better approach than to amalgamate several of my own models? This is the inaugural version, incorporating three of my existing top tier models, the models listed below form the current S͓̽ I͓̽ N͓̽ G͓̽ U͓̽ L͓̽ A͓̽ R͓̽ I͓̽ T͓̽ Y͓̽ checkpoint.

1. ᖻᗩᘻᘿᖇ'S ᕵᘿᖇᖴᘿᑢᖶ ᕲᘿSᓰᘜᘉ - V4 Columna Artis

2. ¥åmêr'§ Äñïmê 🌟💖😏 Úl†rå Ìñ£ïñ - Y's Anime V5

3. 𝚈𝚊𝚖𝚎𝚛'𝚜 𝚁𝚎𝚊𝚕𝚒𝚜𝚝𝚒𝚌 👁‍🗨 𝙽𝚂𝙵𝚆 & 𝚂𝙵𝚆 - V4 + VAE

This model is an experiment for the moment, this model hinges on its success and positive reception. If well-received, I plan to consistently enhance it with more content. The primary focus of this model revolves around various forms of art.

This model is similar to the "Unstable Diffuser" models, it boasts the capability to generate images across a myriad of genres with minimal assistance. While not a top priority, it can also generate content with ease.

Each review accompanied by posted images contributes significantly to improving the model. A heartfelt thank you goes out to all the supporters of my work. I genuinely value and appreciate each review, image, and comment shared on the models I create.

(^_^) The best modifications:

• LoRA: XL YAMER'S STYLE ♠️ Princeps Omnia LoRA

• Restore Face: It is not needed to generate high quality images (Not used in the current showcase.)

• Clip skip: 1 and 2 (The model works well with clip skip set to both.)

• SDXL Refiner: Not needed with my models!

• Checkpoint tested using: A1111
  

🄼 🄾 🄳 🄴 🄻 - 🅁 🄰 🅃 🄸 🄽 🄶:

• Quality: 9/10

• Stability (Variety): 5/10, the lower = the more variety with one prompt

• Flexibility: 8/10

• CFG resistance: 30/30

• Style strength: 5/10 (less=more styles)

• Style Variety: 8.5/10

Where this model is better compared to SDXL 1.0:

• No embedding is needed

• Better variety of style

• No refiner is needed

• Less AI generated look to the image. (When creating realistic images for example)

• Face fix is not needed

• Better saturation, overall quality, texture and tone of the images

• Less blurry images and desaturated content

• No adetailer required (Just use hires fix)

You raise the blade, you make the change.
As the mind goes postal, the art goes flying.
You must have chaos within you to give birth to a dancing star.

-Pink Floyd -Yamer -Friedrich Nietzsche

The recommended settings:

Image Quality: 1024x1024 (Standard for SDXL), 16:9, 4:3, images in the showcase were created using 576x1024

Steps: 35-150 (under 30 steps some artifact may appear and/or weird saturation, for ex: images may look more gritty and desaturated/lacking quality).

• Hires Upscaler: 4xUltraSharp or 4x_foolhardy_Remacri

• Hires upscale: The only limit is your GPU (I upscale 2,5 times the base image, 576x1024)

• VAE: SDXL VAE
  DO NOT POST THIS: Ḑ̷̡̧͓͖̭̠̬̍͛̍̾͛̂ͅO̶̪͖̒͊̒ ̶̺̿͊̈̎͛̂́͠N̶̨͇̫̲̙͍̬̙̹͓̔̀͊͆͂̔͆Ơ̵̲̈͗̏̒͛̄͛̈͘T̵͔̤̮̹͖̹͓̩̣̪͑ ̶̧̰͕̹̹̳̬͕̆͋͐͒̿S̷͈̗̘̣͛̇̄̌̚̚ͅH̵͖̼̑̈́͝Â̴̪̥͈̑̀̈̍̃́͑̿͠R̶̰̝͔͔̘̳͉͚̻̋̀͗̓E̴̹͔͖̜͛͒́̓̓́̿ ̴́̒͊̿̿̂��͍͎̭T̴͓̗͎̉̾̉̑̓̊̑̎̿̐H̸̢̛̲̎́̆̈́̓Ï̴̮̈̉̋̕̕͠S̶̨̘̳̩̮͓̒́̑̊̃̓ ¯\_(ツ)_/¯ you have been warned.

⠀⠀⠀⠀⠀⣠⠀⠀⠀⠀⠀⠀⠀⠀⣀⣀⣀⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠠⡏⠤⠀⠂⠂⣂⣵⣿⣿⣿⣿⣿⣿⣿⣶⣆⠂⠂⠈⠈⠀⠀⠀⠀⠀The singularity has no end.
⠀⠀⠀⠀⠀⠀⠀⠀⢀⣶⣿⣿⡿⠛⠉⠉⠉⠛⢿⣿⣿⣷⣄⡀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⢀⣴⣿⣿⣿⠋⠀⠀⠀⠀⠀⠀⠀⠙⣿⣿⣿⣿⣶⣤⣀⡀⠀⢀       A point in space where nothing matter.
⠀⠀⣀⠰⣀⢿⣿⣿⣿⣿⣿⣀⣀⣀⣀⣀⣀⣰⣰⣶⣿⣿⣿⣿⣿⠿⠿⠁⠉⠁      There must be somebody there, because somebody must have said "Nobody".
⠈⠀⠀⠀⠈⠉⠛⠛⠛⠛⣿⣿⡽⣏⠉⠉⠉⠉⠉⣽⣿⠉⠉⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠻⣷⣦⣄⣀⣠⣴⣾⡿⠃⠀⠀⠀⠀⠀⠀⠀⠀⠀An end to all reason and philosophy, only to leave behind nihilism to fill the gap.
⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⠛⠿⠿⠿⠟⠋⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
                                              One never ending nightmare of void and spirit.