I made an fp8 implementation of flux which gets ~3.5 it/s 1024x1024 on 4090 (ADA / Hopper & 16GB vram+ only)
Open
Feature
- Dominant language
- Python
- Stars
- 133k
- Forks
- 15.7k
- Avg merge
- 1d 6h
- Merged PRs (30d)
- 155
Description
### Feature Idea
Saw the claim on this reddit thread, hopefully the ideas there can also be brought into comfy for even more speedups.
https://www.reddit.com/r/StableDiffusion/comments/1ex64jj/i_made_an_fp8_implementation_of_flux_which_gets/
### Existing Solutions
_No response_
### Other
_No response_
Contributor guide
Assessment
This issue has not been assessed yet.