Containers are the ultimate “works for me” in software development. My experience it makes for more fragile software that depends on its environment being perfect and nothing else will do.
Containers are the ultimate “works for me” in software development. My experience it makes for more fragile software that depends on its environment being perfect and nothing else will do.
I’ve noted over the past few years, how any company that invests in R&D rather than pays dividends is labelled as “loss making” by the press.
No, because I used the web before web search. It’s a convenient thing. Not a necessary thing.
i.e. where it all began
Rocm is the Radeon Open Compute … Something begining with m. It’s the libraries that AMD write to divert workloads to your GPU.
I have a 7600 and I’m finding things a bit bleeding edge (on Linux). I highly recommend making sure you’re on the latest 7.6.1 of Rocm and make sure the version of pytorch you’re using is a recent nightly that uses 7.6. earlier versions didn’t support the 7xxx series.
If you’re on windows, I think thinks can work differently.
Going to share the config?
Sdxl was 3 models too. Base, refiner and vae.
Cascade is 3. Lores, hires, vae.