mtimes — Multiply scalars, vectors, or matrices with matrix-product semantics.
mtimes(A, B) implements the callable form of A * B, including scalar, real, complex, integer-scalar, empty, and provider-resident cases.
Syntax
C = mtimes(A, B)Inputs
A- Type
Any- Required
- Yes
- Description
- Left numeric, logical, character, complex, or provider-resident operand.
B- Type
Any- Required
- Yes
- Description
- Right numeric, logical, character, complex, or provider-resident operand.
Returns
C- Type
NumericArray- Description
- Matrix-arithmetic result.
Errors
RunMat:mtimes:InvalidInput- When
- Operands are unsupported or matrix dimensions are incompatible.
- Message
mtimes: unsupported operand types
RunMat:mtimes:Internal- When
- Runtime cannot complete a provider operation or materialize its output.
- Message
mtimes: internal runtime failure
Matrix product
mtimes(A, B) and A * B multiply scalars, vectors, and matrices. For nonscalar matrices, size(A, 2) must equal size(B, 1). Row-by-column produces a scalar dot product, while column-by-row produces an outer product.
Scalars multiply every element without changing the other operand's shape. Logical and character operands enter floating-point arithmetic. An integer operand is supported only when the other operand is scalar; the result preserves the integer class and saturates.
Complex and empty arrays
Real and complex operands use full complex multiplication when either input is complex. Trailing singleton dimensions are accepted by the runtime's matrix view.
An m-by-0 matrix multiplied by a 0-by-n matrix produces an m-by-n zero matrix. Incompatible inner dimensions return a matrix-dimension error.
Sparse CPU operands
CPU sparse matrix products accept real or complex single and double values, and logical operands enter numeric arithmetic. Two sparse matrix operands produce a numeric sparse result; a sparse matrix paired with a full matrix produces a full result. Scalar products follow the scalar multiplication rules.
Nonscalar sparse matrix products require an unbounded host-memory policy. Calls with finite host-memory quotas are rejected because these sparse output and workspace bounds are not available.
Acceleration
A provider-resident product uses the exact owner of the resident operands. Compatible host operands may be uploaded to that owner. A typed unsupported-operation response permits host fallback; provider failures and invalid output metadata are reported.
Supported provider results retain device residency. Exact integer scalar fallback gathers authoritative typed storage and restores the result to the resident input's provider.
Examples
Multiply two matrices
A = [1 2 3; 4 5 6];
B = [7 8; 9 10; 11 12];
C = A * BExpected output:
C = [58 64; 139 154]Multiply a row by a column
u = [1 2 3];
v = [4; 5; 6];
d = mtimes(u, v)Expected output:
d = 32Scale a matrix
S = 0.5 * eye(3)Expected output:
S = diag([0.5 0.5 0.5])Multiply complex matrices
A = [1+2i 3-4i; 5+6i 7+8i];
B = [1-1i; 2+2i];
C = A * BExpected output:
C = [17-1i; 9+31i]Multiply provider-resident matrices
G1 = gpuArray(single([1 2; 3 4]));
G2 = gpuArray(single([5; 6]));
G = G1 * G2;
R = gather(G)Expected output:
R = single([17; 39])Reject incompatible inner dimensions
A = rand(2, 3);
B = rand(4, 2);
C = A * BExpected output:
Inner matrix dimensions must agree.Using mtimes with coding agents
Open a RunMat example with live inputs, then ask the agent to explain how mtimes changes the result.
Run a small mtimes example, explain the result, then change one input and compare the output.
FAQ
How is mtimes different from times?⌄
mtimes performs matrix products. times, written .*, multiplies corresponding elements with implicit expansion.
Does mtimes support scalars?⌄
Yes. A scalar scales the other operand. Integer inputs require this scalar form and preserve their fixed-width class.
Are complex values supported?⌄
Yes. A real/complex pair is promoted to complex arithmetic.
Will a result stay on its provider?⌄
A valid provider result remains resident. Only a typed unsupported operation may take the documented host fallback path.
Is matmul a callable alias?⌄
No. Use mtimes(A, B) or the A * B operator.
Related Linalg functions
Ops
cross · ctranspose · dot · mldivide · mpower · mrdivide · pagemtimes · pagetranspose · trace · transpose
Structure
bandwidth · isdiag · ishermitian · issymmetric · istril · istriu · symrcm
Factor
chol · decomposition · decomposition.ctranspose · decomposition.mldivide · decomposition.mrdivide · decomposition.mtimes · decomposition.rdivide · decomposition.subsref · decomposition.times · decomposition.uminus · decomposition.uplus · eig · eigs · gmres · ichol · ilu · lu · ordeig · ordqz · ordschur · pcg · qr · qz · rsf2csf · schur · svd
Source you can read
Every RunMat function's source is public. Read exactly how mtimes is executed, line by line, in Rust.
- View the source for mtimes in Rust on GitHub
- Learn how the RunMat runtime works
- Found a bug? Open an issue with a minimal reproduction.
About RunMat
RunMat is a runtime that executes MATLAB-syntax code fast on any GPU. The runtime is free for individuals, including freelancers; students and faculty at an academic institution, for teaching or research there; contributors to an open-source project, for work on that project; organizations with 250 or fewer people and US$10M or less in annual revenue; and anyone, for non-production use. See runmat.com/license.
- RunMat automatically optimizes your math for GPU execution on Apple, Nvidia, and AMD hardware. No code changes needed. Simulations that took hours now take minutes.
- Start running code in seconds. RunMat runs in the browser, on the desktop, or from the CLI. No license server, no IT ticket.