For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Mojo package
amd
AMD GPU convolution kernels.
Architecture-specific implementations live under rdna/ (RDNA 3+).
Packagesβ
- β
rdna: RDNA Conv2D via implicit GEMM (fused im2col + WMMA matmul).
Modulesβ
- β
amd_4wave_conv: 4-wave FP8 implicit-GEMM convolution for AMD MI355X (CDNA4). - β
amd_4wave_conv_residual: AMD 4-wave Conv2D fprop with optional residual add. - β
dispatch: AMD MI355X (CDNA4, gfx950) conv2d dispatch toamd_4wave_conv. - β
dispatch_3d: AMD MI355X (CDNA4, gfx950) conv3d dispatch toamd_4wave_conv.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!