Smaller model size support for Stable Diffusion #502
x64Dev
started this conversation in
Feature Requests
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi first of all great work with Text to image generation. The current stable diffusion models supported to be used as on-device model are approx 4 GB, which makes it impractical to make something as a production, Are there any plan to have small stable diffusion models that can be used in production app keeping in mind the concept behind on-device models.
I'm not sure how people are currently using this in there work specifically as Android apps.
Any suggestion or help is highly appreciated.
Beta Was this translation helpful? Give feedback.
All reactions