For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo function

matmul

def matmul[*, transpose_b: Bool = False, b_packed: Bool = False, elementwise_lambda_fn: Optional[def[dtype: DType, width: SIMDLength, *, alignment: Int = Int(1)](IndexList[Int(2)], SIMD[dtype, width]) capturing thin -> None] = None, saturated_vnni: Bool = False](c: TileTensor[Storage=c.Storage, linear_idx_type=c.linear_idx_type], a: TileTensor[Storage=a.Storage, linear_idx_type=a.linear_idx_type], b: TileTensor[Storage=b.Storage, linear_idx_type=b.linear_idx_type], kernel_type_m: Int, num_threads: Int = Int(-1), ctx: Optional[DeviceContext] = None)

TileTensor matmul dispatcher. Selects kernel type and delegates to _matmul_cpu_impl.

Parameters:

• ​transpose_b (Bool): Whether the B operand is transposed (defaults to False).
• ​b_packed (Bool): Whether B was pre-packed offline in the cache-friendly layout (defaults to False).
• ​elementwise_lambda_fn (Optional[def[dtype: DType, width: SIMDLength, *, alignment: Int = Int(1)](IndexList[Int(2)], SIMD[dtype, width]) capturing thin -> None]): Optional elementwise epilogue applied to each output tile (defaults to None).
• ​saturated_vnni (Bool): When True, uses the saturating x86 variant, which requires the a operand to lie in [0, 127]; it saves instructions only on the AVX2 emulation path used when the target lacks VNNI (defaults to False).

Args:

• ​c (TileTensor[Storage=c.Storage, linear_idx_type=c.linear_idx_type]): Output matrix buffer accumulating the matmul result.
• ​a (TileTensor[Storage=a.Storage, linear_idx_type=a.linear_idx_type]): Left operand of the matmul.
• ​b (TileTensor[Storage=b.Storage, linear_idx_type=b.linear_idx_type]): Right operand of the matmul.
• ​kernel_type_m (Int): M dimension used to select the kernel variant, or 0 if unknown.
• ​num_threads (Int): Number of worker threads to use (defaults to -1, which selects automatically).
• ​ctx (Optional[DeviceContext]): Device context governing parallelism (defaults to None).