llms.txt
Compare with other sites' llms.txt files →# RunMat
RunMat is a math and physics calculation engine designed for broad compatibility with MATLAB syntax and existing `.m` projects. It lets you model, simulate, and analyze complex systems with automatic GPU acceleration. It is available from the CLI, as a collaborative desktop app, and in the web browser.
RunMat is built by Dystr Inc., licensed under Apache 2.0, and written in Rust. It runs as a native application on macOS, Linux, and Windows and in the browser through WebAssembly. Supported GPU execution uses Metal, Vulkan, DirectX 12, or WebGPU depending on the host.
RunMat supports scripts, functions, arrays, control flow, indexing, classes, imports, MATLAB-style packages, private functions, and multi-file projects. It includes more than 1,000 documented built-in functions across numerical computing, data, plotting, signal processing, file I/O, and engineering workflows. Supported MATLAB-language `.m` projects can run without RunMat-specific modifications.
## Start using RunMat
- [Browser Sandbox](https://runmat.com/sandbox): Run MATLAB-syntax code in the browser without installing RunMat or creating an account.
- [Download RunMat](https://runmat.com/download): Install the desktop app or native CLI for macOS, Linux, or Windows.
- [RunMat Desktop](https://runmat.com/docs/desktop): Install Desktop and work with local files, projects, variables, and plots.
- [Command Line Interface](https://runmat.com/docs/runtime/getting-started/cli): Run `.m` files, inspect projects, and use `runmat check` and `runmat test`.
- [GitHub Repository](https://github.com/runmat-org/runmat): Inspect the Apache 2.0 licensed runtime source and tests.
- [Changelog](https://runmat.com/docs/changelog): Review current releases and product changes.
## Authoritative documentation
- [MATLAB Compatibility](https://runmat.com/docs/runtime/matlab-compatibility): Language semantics, indexing, classes, packages, compatibility modes, project validation, supported built-ins, and current boundaries.
- [Built-in Function Reference](https://runmat.com/docs/reference/builtins): Search more than 1,000 documented built-in functions, including syntax, inputs, outputs, errors, examples, GPU behavior, and source links.
- [Projects and `runmat.toml`](https://runmat.com/docs/runtime/getting-started/projects): Organize source roots, local dependencies, packages, classes, private functions, test directories, and named entrypoints.
- [GPU Acceleration](https://runmat.com/docs/runtime/gpu): Understand automatic CPU/GPU execution, fusion, provider support, and data residency.
- [Testing Strategy](https://runmat.com/docs/runtime/development/testing): Review RunMat's broader automated test suite, compatibility testing, and CPU/GPU parity checks.
## Compatibility workflow
Use the real project and the current RunMat release for compatibility validation:
```bash
runmat check analysis.m
runmat run analysis.m
runmat test
```
`runmat check` analyzes syntax, semantics, source resolution, and proven type or shape incompatibilities without executing the project. `runmat run` exercises runtime behavior and external data. `runmat test` discovers and runs supported MATLAB-style script, function, and class tests.
Compatibility depends on the exact language features, function call forms, toolboxes, file formats, native extensions, external products, and hardware integrations used by a project. Simulink block diagrams and GUIDE or App Designer graphical applications are separate workflows.
## High-intent guides and comparisons
- [RunMat as a MATLAB Online Alternative](https://runmat.com/matlab-online): Compare browser execution, local files, GPU access, collaboration, and offline use.
- [Free MATLAB Alternatives](https://runmat.com/blog/free-matlab-alternatives): Compare RunMat, GNU Octave, Python, Julia, and other options.
- [Using a GPU with MATLAB-language Code](https://runmat.com/blog/how-to-use-gpu-in-matlab): Compare explicit MATLAB GPU workflows with RunMat's automatic cross-vendor execution.
- [Benchmarks](https://runmat.com/benchmarks): Review measured workloads, hardware, methods, and results.
## Company and legal
- [About RunMat](https://runmat.com/about): RunMat's company, product, and mission.
- [License](https://runmat.com/license): Licensing information for the open-source runtime and hosted product.
- [Contact](https://runmat.com/contact): Product and company enquiries.
RunMat is an independent runtime. MATLAB is a registered trademark of The MathWorks, Inc. RunMat is not affiliated with, endorsed by, or sponsored by The MathWorks, Inc.