FeenoX
seamplex.com
- Free plan
- Linux
- Web
- API
- Self-hosted
What FeenoX is
FeenoX is a free, open-source computational engineering tool for advanced users working on GNU/Linux. It handles dynamical systems described by ODEs or DAEs and finite-element PDEs, including heat conduction, thermo-mechanical problems, modal analysis, neutron diffusion, and neutron transport. Users describe problems in English-like plain-text input files, so they can change a model without recompiling. FeenoX reads Gmsh .msh meshes, uses GNU GSL, SUNDIALS, PETSc, and SLEPc for computation, and can produce Gmsh MSH or ParaView VTK/VTU results. It supports MPI parallelization and is designed for remote servers. Installation options include Debian or Ubuntu packages, source tarballs, GitHub, and generic statically linked GNU/Linux binaries. Ubuntu availability starts with version 25.04. Those generic binaries omit MUMPS and MPI and use basic optimization flags. FeenoX has no graphical interface, and the documentation targets GNU/Linux rather than Windows. Browser access is available through interfaces such as SunCAE and CAEplex.
Who it suits
FeenoX suits engineers, Unix users, and academics comfortable with advanced GNU/Linux workflows and text-based problem definitions. Its API-oriented design and MPI support also fit projects integrating computational engineering into web interfaces or remote-server workflows.
Good to know, and what to watch
- Free and open source under GNU GPLv3 or later
- Supports ODEs, DAEs, and finite-element PDEs
- MPI parallelization for multi-computer computation
- Exports results in Gmsh and ParaView formats
- Plain-text inputs can be changed without recompiling
- No graphical interface
- Documentation targets GNU/Linux, not Windows
- Generic binaries omit MUMPS and MPI
Plans
Every plan and priceFeenoX
Free
Free and no-fee; no paid pricing is stated.
Free plan
Getting started
- Open the FeenoX website
- Choose Debian or Ubuntu packages, static GNU/Linux binaries, source tarballs, or the GitHub repository
- Install on a supported GNU/Linux system
- Create a self-descriptive plain-text problem file
- Provide meshes in Gmsh .msh format when needed
- Run the model and export Gmsh or ParaView results
Limits to check
The generic statically linked GNU/Linux binaries do not support MUMPS or MPI and use only basic optimization flags. FeenoX has no graphical interface and its documentation targets GNU/Linux.
The Club's review
FeenoX is a free, open-source computational engineering solver for people who are comfortable expressing analyses in text and working in GNU/Linux-oriented environments. It is best suited to engineers, academics and Unix-minded developers building repeatable or automated workflows rather than to users shopping for a visual desktop modeller.
Overview
FeenoX addresses a wide span of engineering computation: mathematical problems, ordinary and differential-algebraic equations, heat conduction, thermo-mechanical analysis, modal analysis, neutron diffusion and neutron transport. You describe each problem in a self-descriptive, English-like plain-text file, then change that file when the model changes instead of recompiling a program. That is a strong basis for version control, scripting and reproducible studies, but it assumes you are willing to author and maintain technical input files.
The software is free in both price and freedom. It is released under the GNU General Public License version 3 or any later version, allowing use, modification and redistribution under the license terms. Source code and dependencies are open, and the project includes unit, regression, mathematical and code-verification tests. The license carries no warranty to the extent permitted by law, so organisations need their own engineering assurance and support processes.
Key features
Text-driven models and established solvers
Meshes are read in Gmsh .msh format, while results can be written as Gmsh MSH or ParaView VTK/VTU files. GNU GSL, SUNDIALS, PETSc and SLEPc provide the computational building blocks. This is a practical choice for teams already using those ecosystems and for pipelines that need standard post-processing; it is less attractive if your workflow depends on a built-in model editor or proprietary project format.
Parallel and cloud-oriented operation
MPI parallelization lets suitable installations spread work across multiple computers, and the design targets arbitrarily scalable remote servers. FeenoX can be integrated with web interfaces, REST APIs, large language models, Markdown and LaTeX reporting, and standard post-processing tools. Web access is available through interfaces such as SunCAE and the browser-based CAEplex service. Those integration points make it credible as a backend for an engineering service, while adding setup and infrastructure responsibilities that a packaged desktop application would normally hide.
Open development
Users can seek help in GitHub Discussions and report bugs through GitHub Issues. Contributions from hackers and academics are welcomed, particularly new partial differential equations and formulations. That openness benefits organisations able to inspect or extend the implementation, but it does not amount to a bundled commercial support contract.
Pricing
FeenoX: free, billed as GNU GPLv3+ open-source software. There is no subscription charge, seat fee or usage quota. The free offering includes the complete project, but no graphical interface, targets GNU/Linux, and the generic binaries omit MUMPS and MPI.
Free: free. This is the no-fee option rather than a limited trial; no paid pricing is stated. Choose it when you can work within the GNU/Linux and command-line model. Users who need MPI or MUMPS must use an installation route that supports them rather than relying on the generic binaries.
Platforms
FeenoX is delivered for API, Linux, self-hosted and web use, with offline access, data export, authoring tools and 3D visuals. Official installation choices include Debian or Ubuntu packages, statically linked GNU/Linux binaries, source tarballs and the GitHub repository. Ubuntu support starts with Ubuntu 25.04. The documentation targets GNU/Linux, the project has no graphical interface, and it is not expected to run in Windows. Generic static binaries are convenient to deploy but do not include MUMPS or MPI and use only basic optimization flags.
Who it's for
Choose FeenoX if you are an industry engineer, academic professor or Unix hacker who needs scriptable multiphysics calculations, transparent source and the ability to connect a solver to remote compute or web services. It is particularly well matched to repeatable studies, research code and organisations willing to manage meshes, inputs, post-processing and Linux deployment themselves. It is the wrong fit for beginners seeking visual setup, for Windows-first teams, or for departments that require a vendor-backed warranty and turnkey desktop experience.
Pros and cons
- Pros: broad coverage from ODE/DAE systems to thermal, structural and nuclear analyses; plain-text inputs support automation and reproducibility; MPI and API-oriented design can feed multi-computer or web workflows; GPL licensing, open dependencies and verification tests support inspection and extension; Gmsh and ParaView formats simplify interchange.
- Cons: no GUI means model creation is entirely text- and tool-driven; official focus on GNU/Linux excludes a normal Windows desktop path; generic binaries lack MUMPS and MPI, limiting out-of-the-box parallel performance; users must arrange their own support and accept the license's absence of warranty.
Alternatives
For a browser-friendly or graphical workflow across Windows and macOS, SimFlow is the better alternative: it has Linux, macOS, web and Windows support, a free plan and a free trial. Teams needing a commercial multiphysics product with Windows and macOS support should consider COMSOL Multiphysics, which is paid and offers a free trial.
Xcos is a free choice for users wanting a block-diagram environment distributed with Scilab. For command-line or file-based circuit simulation without schematic entry, choose ngspice. OpenFOAM is another free, GPL-based option when its CFD ecosystem is the priority, while support and code customisation are separate services. OpenSim suits users seeking a free open-source platform available through API and desktop operating systems. MPTRAC offers open-source Linux tooling with a browser-based Web Runner. FEST-3D is a free GPL-licensed alternative for Linux and macOS. Browse the Physics Simulation Software directory for the wider category.
Verdict
FeenoX is a strong recommendation for technically advanced Linux users who value open code, editable text inputs and scalable, integrable engineering computation at zero license cost. Look elsewhere if your priority is a graphical interface, native Windows operation, bundled enterprise support or generic binaries with MPI and MUMPS ready to run.
The verdict
FeenoX offers a broad set of engineering computations in a text-driven, open-source package. Its lack of a GUI and the limits of its generic binaries make it a better fit for technically advanced users than those seeking a desktop application.
Questions people ask
Is FeenoX free?
Yes. FeenoX is free in both the no-cost and freedom senses, and the listed FeenoX plan is free under GNU GPLv3 or later.
Which systems does FeenoX support?
Official support targets GNU/Linux, including Ubuntu 25.04 onward. It is not expected to run in Windows.
Does FeenoX have a graphical interface?
No. Browser-based access is available through interfaces such as SunCAE and CAEplex.
What file formats does FeenoX use?
Input meshes use Gmsh’s .msh format. Results can be written as Gmsh MSH or ParaView VTK/VTU files.
Can FeenoX run in parallel?
Yes. It supports MPI parallelization and is designed for arbitrarily scalable remote servers.
What the maker publishes
- API integration
- The design supports straightforward integration with web interfaces, large language models, REST APIs, standard post-processing tools, Markdown and LaTeX reporting. source
- Assessment tools
- No source
- Audience
- Official pages present FeenoX for industry engineers, Unix hackers and academic professors, with advanced users as the target audience. source
- Authoring tools
- Yes source
- Binary limitation
- The generic statically linked GNU/Linux binaries do not support MUMPS or MPI and use only basic optimization flags. source
- Binary limitations
- Generic binaries do not support MUMPS or MPI and use only basic optimization flags. source
- Contributions welcome
- Contributions from hackers and academics are welcome, especially new PDEs and formulations. source
- Data export
- Yes source
- Delivery mode
- desktop source
- Free open source
- FeenoX is free in both the no-cost and freedom senses, and open source. source
- Free plan
- Yes source
- GitHub support
- Users can ask for help on the GitHub discussions page. source
- GUI limitation
- FeenoX has no graphical interface and is not expected to run in Windows; the official documentation targets GNU/Linux. source
- Plain-text inputs
- Problems use syntactically sugared, self-descriptive English-like plain-text input files. source
- Input model
- Problems are defined in self-descriptive English-like plain-text input files, so changing a problem does not require recompiling. source
- Installation
- Official installation options include Debian or Ubuntu packages, statically linked GNU/Linux binaries, source tarballs and the GitHub repository. source
- Integrations
- It reads Gmsh meshes, uses PETSc and SLEPc solvers, and writes Gmsh MSH or ParaView VTK/VTU results. source
- Lesson library
- No source
- License
- FeenoX is distributed under the GNU General Public License version 3 or any later version, with freedom to use, modify and redistribute it under the license terms. source
- GNU/Linux binaries
- Generic GNU/Linux binaries are provided as statically linked executables. source
- Gmsh mesh format
- Input meshes use Gmsh's .msh format. source
- MPI parallelization
- FeenoX supports MPI parallelization. source
- Offline access
- Yes source
- Parallelism
- FeenoX supports MPI parallelization and is designed for arbitrarily scalable remote servers. source
Checked against the maker's pages on 28 Sept 2026.