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 module

impl

Implements tiled CPU matrix multiplication kernels and their dispatcher.

Defines the InnerMatmulKernel trait and the TiledMatmul struct that drive the outer M/N/K tile loops, plus the matmul entry point that selects an inner microkernel (default, VNNI, NEON, or I8MM) or routes to Apple Accelerate and GEMV fast paths.

Structs

• ​TiledMatmul: Tiled matmul implementation integrating packing, inner loop and tile partitions.

Traits

• ​InnerMatmulKernel: Trait for CPU matmul microkernels operating on pre-packed tiles.

Functions

• ​elementwise_epilogue_c_tile: Applies a vectorized epilogue function over a 2D C output tile.
• ​matmul: TileTensor matmul dispatcher. Selects kernel type and delegates to _matmul_cpu_impl.
• ​tiled_matmul_run: Interface function to run tiled matmul on a given sub-tile.

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!

😔 What went wrong?

Some code doesn’t work

It includes inaccurate information

It's missing information I need

It was difficult to understand

Other

Submit