https://huggingface.co/stabilityai/sd-vae-ft-mse-original/blob/main/vae-ft-mse-840000-ema-pruned.ckpt https://huggingface.co/AIARTCHAN/aichan_blend/tree/main/vae Apply VAE. You will get better color results.
A: This model has a painting style of 2.2d. Like DosMix, the iris of the pupil is clear.https://civarchive.com/models/6250/dosmix
Sampler tested: eluer a, DPM++SDE Karras)
Recommended: step 20, scale 6-11 hires fix denoise 0.5, upscale by 2. Latent, R-ESRGAN 4×+ Anime6B.
Don't tex2img, try img2img SD upscaling. That's a good way too.
img2img SD upscale method: scale 20-25, denoising 0.2-0.3 After selecting SD Upscale at the bottom, tile overlap 64, scale factor2
caution! Sampler must be DPM++SDE karras.
It is recommended, not obligatory.
Recommendation Prompt: Best Quality, Masterpiece, Negative prompt: (low quality, worst quality: 1.4) This is a recommendation, not a compulsion. You guys make me better than me.
A higher scale will give you beautiful results, but giving the AI too many degrees of freedom will make the prompts irrelevant.
Description
A
FAQ
Comments (11)
Pls dont use tags that dont apply like photorealistic tag.
ok!
I suggest naming models with a alphabet suffix or codename instead of version number if they are alternative versions instead of upgraded versions
Is it possible to explain how you trained this model? or any tutorial?
Hello, very good. Your model wanted to know how you do it. Do you have a link where you learned from? Thank you very much.
May I ask you to make a .ckpt version with a baked VAE? I can only use DiffusionBee App on Intel Mac but only .ckpt understands it and can't use external VAE
have you tried Draw Things, it does ckpt, safetensor and VAE, it also does control net, in edge detect, depth map and scribble. It doesn't do LoRa yet but does Texual Inversions. It also integrates with Civitai, you can click the "play" on the download link ant it will send it to Draw Things.
What happened with the other versions
The ambulance was split and disembarked. I forgot the merge ratio even though I want to recover it... sorry..
I saw your Cartoonish model .. Wasn't able to download it before take down..
Can you provide mirror link?
uploaded now