Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Also good news was that running olive and downloading the ONNX converter to run stable diffusion shows something like 120% speedup on Nvidia and 50% speedup on AMD from the numbers I see. Some UIs are beginning to support ONNX format. ONNX being the format that has Microsoft's support behind it for AI models to my understanding.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: