About samplers #102
Unanswered
tetsuoo-online
asked this question in
Q&A
Replies: 1 comment
-
This is probably because of changes to the model in InvokeAI, the fork of stable diffusion this addon uses as a submodule. So it isn’t 100% compatible with other forks. FYI, all samplers should eventually resolve to the same image given enough steps. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi, I'm posting here because I'm not sure it's an issue, could be the user :)
I just noticed the samplers so I was playing around a bit, then... I decided to replicate the same generated image in Colab https://colab.research.google.com/drive/1jUwJ0owjigpG-9m6AI_wEStwimisUE17
Using the same model, same seed, steps etc I finally got about 99% the same result :
Now assuming samplers and schedulers are the same thing (right ?), I'm wondering which one is wrong ? x)
If I use DDIM in Colab I will get a completely different result
Beta Was this translation helpful? Give feedback.
All reactions