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 function
multistage_gemm
def multistage_gemm[c_type: DType, a_type: DType, b_type: DType, //, *, transpose_b: Bool, config: MatmulConfig[a_type, b_type, c_type, transpose_b], elementwise_lambda_fn: Optional[def[dtype: DType, width: SIMDSize, *, alignment: Int = Int(1)](IndexList[Int(2)], SIMD[dtype, width]) capturing -> None] = None](c: TileTensor[c_type, Storage=c.Storage, address_space=c.address_space, linear_idx_type=c.linear_idx_type, element_size=c.element_size], a: TileTensor[a_type, Storage=a.Storage, address_space=a.address_space, linear_idx_type=a.linear_idx_type, element_size=a.element_size], b: TileTensor[b_type, Storage=b.Storage, address_space=b.address_space, linear_idx_type=b.linear_idx_type, element_size=b.element_size], ctx: DeviceContext)
TileTensor overload of multistage_gemm. Converts to LayoutTensor and dispatches to the appropriate GEMM kernel.
def multistage_gemm[c_type: DType, a_type: DType, b_type: DType, //, *, transpose_b: Bool, config: MatmulConfig[a_type, b_type, c_type, transpose_b], elementwise_lambda_fn: Optional[def[dtype: DType, width: SIMDSize, *, alignment: Int = Int(1)](IndexList[Int(2)], SIMD[dtype, width]) capturing -> None] = None](c: TileTensor[c_type, Storage=c.Storage, address_space=c.address_space, linear_idx_type=c.linear_idx_type, element_size=c.element_size], a: TileTensor[a_type, Storage=a.Storage, address_space=a.address_space, linear_idx_type=a.linear_idx_type, element_size=a.element_size], b: TileTensor[b_type, Storage=b.Storage, address_space=b.address_space, linear_idx_type=b.linear_idx_type, element_size=b.element_size], runtime_config: MatmulConfig[a_type, b_type, c_type, transpose_b], ctx: DeviceContext)
TileTensor overload of multistage_gemm with runtime config. Constrains c to mut=True because split_k_reduce requires a mutable output tensor.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!