faster inference
rom1504 opened this issue · comments
Romain Beaumont commented
using caching @borzunov implemented 10x faster generation at https://github.com/learning-at-home/dalle-pytorch/pull/3/files
I think this could be useful
Phil Wang commented
oh nice! yeah, I can get this done for both dalle and nuwa in one go when I get a free stretch of time
Alexander Borzunov commented
Hey! I'll make a PR to this repo with a finished version of this code today or tomorrow :)
Phil Wang commented
🙏 🙏
Alexander Borzunov commented