Welcome to AzureBlue!
I developed this AI model with the goal of leveraging SDXL to generate high-quality anime images.
For optimal results, I recommend using the DPM++ 2SA Karras sampler with 16 to 30 steps.
AzureBlue was crafted by merging several highly regarded models:
Swim In Latent
(https://civarchive.com/models/118525/swim-in-latent) by @ddPn08
CounterfeitXL
(https://civarchive.com/models/118406/counterfeitxl) by @rqdwdw
ShikiAnimeXL
(https://civarchive.com/models/119163/shikianimexl) by @4444zyf696
animagine-xl
(https://huggingface.co/Linaqruf/animagine-xl/tree/main) by @Linaqruf
Each model contributes unique characteristics to AzureBlue, creating a balanced blend that offers versatility, precision, and aesthetic appeal.
Remember, as with any tool, the results you achieve with AzureBlue can vary based on your input and settings. I encourage you to experiment with different settings and inputs to find what works best for your creative needs.
Your likes, reviews, and constructive feedback are invaluable to me as I continue to refine and improve AzureBlue. If you find AzureBlue useful, don't hesitate to show your support by liking and reviewing it.
Happy generating with AzureBlue!
Description
FAQ
Comments (2)
Something went wrong. The faces suffering some kind of blur and "doublecontouring"!
I'm not yet fully familiar with the issues arising from SDXL, and I'm not sure if this will be the right solution for you, but it seems that with certain samplers, the rendering can break down. I suggest trying DPM++ 2S a Karras to see if it helps.