gt — Compare values with the element-wise > relation.
gt(A,B) and A > B compare compatible operands element by element or construct symbolic relations.
Syntax
tf = gt(A, B)Inputs
A- Type
Any- Required
- Yes
- Description
- Left operand.
B- Type
Any- Required
- Yes
- Description
- Right operand.
Returns
tf- Type
LogicalArray- Description
- Logical greater-than result, or a symbolic relation for symbolic operands.
Errors
RunMat:gt:InvalidInput- When
- Operands contain unsupported types or mix numeric and text domains.
- Message
gt: mixing numeric and string inputs is not supported
RunMat:gt:SizeMismatch- When
- Operands are not broadcast-compatible.
- Message
gt: array sizes are not compatible for broadcasting
RunMat:gpu:ProviderOwnershipMismatch- When
- Resident operands do not have one exact owning provider.
- Message
gt: resident operands must have one exact owning provider
RunMat:gt:GpuUploadFailed- When
- An explicitly resident logical result cannot be restored to its provider.
- Message
gt: failed to preserve explicit gpuArray residency
Comparison semantics
gt(A,B) and A > B compare corresponding elements after implicit expansion and return where the left operand is greater than the right operand. Numeric, logical, character, string, categorical, complex, and symbolic operands follow their domain rules.
Ordered comparisons project complex numeric values onto their real components. Any comparison involving NaN is false. Fixed-width integers retain exact values across signed, unsigned, and floating operands; character data compares by code point against numeric data and as text against string data.
Symbolic and accelerated execution
A symbolic operand produces a symbolic relation with the broadcast result shape. Nonsymbolic comparisons produce logical values.
Compatible resident operands use the exact provider's ordering operation. Complex-interleaved inputs project validated real lanes before comparison. Unsupported routes use the exact host implementation and restore explicitly requested residency. Compatible elementwise expressions may fuse.
Examples
Compare scalars
tf = gt(42, 17)Expected output:
tf = trueCompare a matrix with a threshold
A = [1 2 3; 4 5 6];
tf = gt(A, 3)Expected output:
[0 0 0; 1 1 1]Use implicit expansion
tf = gt([1; 3], [2 3 4])Expected output:
[0 0 0; 1 0 0]Compare character code points
tf = gt(['A' 'B' 'C'], 66)Expected output:
[0 0 1]Compare strings lexicographically
tf = gt(["alice" "charlie" "bob"], "bob")Expected output:
[0 1 0]Order complex values by real component
tf = gt([1+99i 3-99i], 2)Expected output:
tf matches the comparison of [1 3] with 2Keep an ordering result resident
A = gpuArray([1 4 7]);
B = gpuArray([0 5 6]);
gtf = gt(A, B);
tf = gather(gtf)Expected output:
[1 0 1]Using gt with coding agents
Open a RunMat example with live inputs, then ask the agent to explain how gt changes the result.
Run a small gt example, explain the result, then change one input and compare the output.
FAQ
What class does gt return?⌄
Nonsymbolic comparisons return logical values. Symbolic operands construct symbolic relations.
How are NaN values treated?⌄
Every ordered comparison involving NaN is false.
How are complex values ordered?⌄
Only the real components participate in ordering.
How are strings compared?⌄
String values compare lexicographically, with implicit expansion for compatible shapes.
How are character arrays compared?⌄
Against numeric data they compare as code points; against text they compare as text.
Are wide integers rounded?⌄
No. Mixed integer comparisons retain exact fixed-width values.
Does a GPU comparison gather?⌄
A supported same-owner operation stays resident. Exact fallback restores explicitly requested residency.
Can the comparison fuse?⌄
Compatible elementwise regions may fuse on the selected provider.
Related Logical functions
Source you can read
Every RunMat function's source is public. Read exactly how gt is executed, line by line, in Rust.
- View the source for gt 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.