▼ Recommended Settings ▼

Negative Prompt
Sampling Method
DPM++ SDE Karras
Sampling Steps
12 (8 ≈ 16)
Restore Faces
Off
Hires Fix (! Required !)
On
Upscaler: Latent (bicubic antialised)
Hires Steps: 5 (4 ≈ 10)
Denoising Strenght: 0.55 (0.4 ≈ 0.7)
Upscale by: 2
CFG Scale
3 (2 ≈ 5)
Clip Skip
1 (1 ≈ 2)
VAE
None (Baked)
▼ Merge Based On ▼
▼ More Info ▼
Any questions or suggestions you can contact me through Discord Cisney Gassai#5108 or Twitter @Realisian.
For donations or requests you can do it on my Ko-fi page <3.
Description
Weightings adjusted
Better Faces
Better Skin
Added Pruned FP32 and FP16 models
FAQ
Comments (18)
This model should be called MakimaBased lmao
Haha sorry, you don't need it but I used it for the V1 and I though I should keep using it for the main showcase images :).
By comparison, I feel that version 4 tends to be more responsive to prompts while maintaining the face and body, It's better expressed if the character's personality or other subtle expressions are described in the prompt
Is there an inpainting version? or do you use a different model for that? sorry im still new to all this
No and I really haven't tested the model with inpainting so I don't now how it does.
Hey. Look up online but you can make any model into an inpainting model. You just have to merge it with the 1.5 SD Inpainting model using certain weight values
@angryliberal009325 nice, thanks. About certain weights you mean Lora or sth else?
why the picture get blurry when denoising strength is set below 0.5
That's normal, latent upscaling do require more denoising, there is a way to fix that and be able to use lower values but introduces other problems so you better stick with the suggested value or use another upscaler.
Where can I found this? <lora:makimaChainsawMan_v10:0.4>
@Cisney_Gassai thank you:)
dayum how do they (gallery below) make gif? is it deforum or what?
"half body" "full body" seems to be hard to work with it.
Try putting "Close up, portrait" in Negative
Amazing work! Are there any upcoming updates? I'm looking forward to them.
Thanks, I'm working on a V5 but I haven't had enough time lately so I'm not sure when I'll release it.
Details
Files
realisian_v40.safetensors
Mirrors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40_1.safetensors
Realisian_v4.safetensors
realisian_v40.safetensors
Realisian.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
291_realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realbilek.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
realisian_v40.safetensors
Realisian.safetensors







