Update app.py to 8 steps and 1.2 LoRA weight as default

#1
by AP123 - opened

Improve default settings for demo generation

Increased the number of inference steps from 4 to 8 and adjusted the default LoRA weight from 0.8 to 1.2. These changes yield better initial results, which should encourage adoption and further development by providing a stronger baseline for new users demoing LCMs for the first time!

multimodalart changed pull request status to merged

Sign up or log in to comment