moe-hip / README.md
pat883's picture
Add kernel card (library_name: kernels)
cc9800c verified
|
Raw
History Blame Contribute Delete
560 Bytes
metadata
library_name: kernels
tags:
  - kernels
  - rocm

MoE Token Alignment (RDNA4/gfx1201)

MoE block-aligned expert-sort (moe_align) native HIP kernel.

Built with kernel-builder for AMD RDNA4 (gfx1201). Load with the kernels library:

from kernels import get_kernel
kernel = get_kernel("pat883/moe-hip")

Requires a ROCm PyTorch build (torch 2.10 / ROCm 7.x) on an RDNA4 card. Built variants: torch210-cxx11-rocm70 and torch210-cxx11-rocm71.