So it has been a bit for this checkpoint, but version 3 is finally ready for some sharing. The images do not come across as soft and the VAE that is baked in is: https://huggingface.co/madebyollin/sdxl-vae-fp16-fix
I decided to use this VAE because it seems to do better and run faster. Overall I have enjoyed the results. I hope you do as well.
Input
Per image generation will cost 0.0047$
For premium plan image generation will cost 0.00$ i.e Free.
Output
Unknown content type
Related Models
Discover similar models you might be interested in
NN2 - NNKCSV1
Andrea - V1
Hubble Telescope LoRA - Beta
J3nniLpz - v1.0
CLO_pleated_skirt_vintage_V1 - V1
Columbina - v1.0-SDXL
myads_desi_lora_genz - V1
Steamed - V1
gothluna - V1
Nahida V1.0 (LoRA on AnimagineXL4.0) - v1.0
BASKET-PACK SDXL Medieval Style Backpack - V1
KIYOI - V1