Downloads · 30 days
2
7% of all-time downloads
w4a8-fp8-wmma
pat883/w4a8-fp8-wmma
w4a8-fp8-wmma is a machine learning model from pat883. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for kernels.
W4A8-FP8 WMMA dense GEMM and MoE GEMM (fat binary for gfx1200 + gfx1201).
All-time downloads
28
Public
Repo size
29.5 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.so29.5 MB · 100%
At a glance
- Library
- kernels
- Access
- Public
- Created
- Jul 3, 2026
- Updated
- Jul 3, 2026
- SHA
- 9fc59f8e
Model card
From the Hugging Face model README
W4A8-FP8 WMMA Dense + MoE GEMM (RDNA4/gfx1200+gfx1201)
W4A8-FP8 WMMA dense GEMM and MoE GEMM (fat binary for gfx1200 + gfx1201).
Built with kernel-builder for AMD RDNA4 (gfx1201).
Load with the kernels library:
from kernels import get_kernel
kernel = get_kernel("pat883/w4a8-fp8-wmma")
Requires a ROCm PyTorch build (torch 2.10 / ROCm 7.x) on an RDNA4 card. Built variants:
torch210-cxx11-rocm70andtorch210-cxx11-rocm71.
- Library
- kernels
- Created
- Jul 3, 2026
- Updated
- Jul 3, 2026