Repository navigation
Replies: 1 comment 1 reply
|
Thanks a lot, and sorry for the slow reply! Yes: the prompt enhancer is in the works for 0.10.0, together with LoRA plugins. It runs on the Qwen3-VL model Kokon already loads as the text encoder, so there's no extra download and no second model in memory, and you'll see your prompt being rewritten live before the image starts. It's optional (a toggle next to the prompt), and you can edit its instructions in Settings. I'll post here when it's out. |
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
This is very nice. Thx for OS this. Faster than mine that i build. I will berry mine and use yours :D
Are you planning to add prompt enhancer?
All reactions