Scale (run CUDA on AMD GPUs without mods) supports gfx900 and gfx1102
What would differentiate this from AMD’s clang based HIP compiler https://rocm.docs.amd.com/projects/HIP/en/latest/ ?
undefined
What would differentiate this from AMD’s clang based HIP compiler https://rocm.docs.amd.com/projects/HIP/en/latest/ ?
undefined