Zero123++
#2421
Replies: 2 comments 1 reply
-
Bit updated steps: Prepare:
Activate:
Deactivate:
|
Beta Was this translation helpful? Give feedback.
0 replies
-
Upd. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
For Zero123++;
Set "Execution backend" to "diffusers" and fully restart.
Set "Custom Diffusers pipeline to use" to
Disty0/zero123plus-pipeline
Set "Diffusers pipeline" to "Custom Diffusers Pipeline".
Then download
sudo-ai/zero123plus-v1.1
from Models/Huggingface tab.(Enter
Disty0/zero123plus-pipeline
as the custom pipeline in Models/Huggingface tab too.)Use Img2Img tab with these settings:
Sampler: DEIS or DPM++ 2M (With Karras)
Steps: 75 (For anime. Realistic can use less steps.)
CFG: 4
Image CFG: 1
You can use simple prompts but detailing it too much can cause nightmare fuels.
Negatives will be ignored.
Beta Was this translation helpful? Give feedback.
All reactions