Disty0

Using TCMalloc fixes it but Intel Compute Runtime can fail with TCMalloc or Jemalloc on some systems. Edit: TCMalloc doesn't completely fix it but definitely helps to reduce it. Still...

> Hi @Disty0 , thanks for sharing this finding. Could you help us to reproduce your execution on our side? Any guidance or commands would be helpful. Code structure of...

Does this still happen in latest version? Quanto changed its api so we had to change our hijcaks to make qint4 load without taking ages.

Hanging issue should be fixed. > I don't understand why this line says device=torch.device("cpu") (why cpu?) Using quantization means you don't have enough VRAM to store the model in GPU,...

Model caching reduces the startup time from minutes to a few seconds if a cache for the model already exists. There is a wiki page for Model Caching: https://github.com/vladmandic/automatic/wiki/OpenVINO#model-caching

Set to true on dev branch: https://github.com/vladmandic/automatic/commit/6d3d23bddd6efd53196a400795ccfcdc31e6723b

> Hi [@Disty0](https://github.com/Disty0) , we don't have OS Arch Linux support. May I know how you install the driver on Arch Linux? intel-compute-runtime-bin package takes the .deb files from the...

I can reproduce the exact same issue on Ubuntu 25.04 / plucky with glibc 2.41. Ubuntu noble with glibc 2.39 works fine. Run Ubuntu plucky in docker: ```sh docker run...

Changed the issue title from Arch Linux to GLIBC 2.41 as it is reproducible on other distros with GLIBC 2.41 too.

IPEX 2.7 has this issue too.