Getting started with FERS
FERS is a finite-element solver for beams, frames and trusses with EN 1993-1-1 steel checks. The same Rust engine runs in your browser, in a Python script, in a Node process and behind the REST API — so you can start with no install at all and move to code whenever it pays off.
What is FERS?
FERS pairs a Rust solver with a Python modelling layer and a browser workspace. You decide where each step runs:
- Local only — install the package (
pip install FERS, imported asfers_core), define the problem in Python, solve on your machine, export structured results. No account, no network. - Browser only — build the model in the web workspace and solve it client-side in WebAssembly. Nothing is uploaded, and no account is needed to get an answer.
- Hybrid — prototype in a script, then push the model to FERS Cloud for visualization, sharing and 3D export; or build it in the browser and export the JSON into a script.
All four routes consume the same model JSON, in SI base units, so a model started one way can always be finished another.
Documentation
Each page below is a complete, self-contained topic.
Quickstart: your first model in the browser
Build and solve a 3D frame in the browser: materials, sections, nodes, supports, members, loads, Calculate.
Install the FERS Python package
Install with pip, verify the import, and set up optional cloud authentication with an API key or token.
Build and solve a model in Python
The core Python objects, how they fit together, and how to read displacements, internal forces and reactions.
Worked examples
Four runnable scripts — cantilever, simply supported beam, portal frame and an EN 1993-1-1 check — each with its hand check.
Use FERS from JavaScript
Install the WebAssembly solver from npm, configure your bundler, and solve models in the browser or in Node.
REST API reference
Every SDK endpoint with its request and response payload — auth, keys, solve, validate, beam check and model CRUD.
Keyboard shortcuts
Every shortcut and mouse gesture in the modelling workspace — tools, editing, selection and view.
Pick your route
Same engine, same results, four ways in.
Python
Build models as objects, sweep parameters in a loop, and get results back as data. The route for anything repetitive or that needs to live in version control.
Structural analysis in PythonJavaScript
A WebAssembly build of the solver, published on npm. Runs in the browser or Node with no server round-trip, so the model never leaves the user's machine.
Structural analysis in JavaScriptREST API
Plain HTTP from any language, with an OpenAPI 3.1 spec you can import straight into a client generator or a ChatGPT Custom GPT Action.
REST API referenceFor AI agents
FERS has a remote MCP server, so Claude, ChatGPT, Cursor and VS Code can build and solve structural models directly — including EN 1993-1-1 member checks with a per-clause trace. Connect it once and your assistant calculates rather than guesses.
Agents that would rather read than call: llms.txt indexes this site, llms-full.txt is the whole documentation in one fetch, and every page here has a plain-markdown twin at /docs/<topic>.md.
Getting help
Resources for troubleshooting and support.
Frequently asked questions
Do I need to install anything to try FERS?
What does the free tier include?
Is my model uploaded to your servers?
How accurate is it?
Which units and axis conventions does FERS use?
i_z governs ordinary bending. Full details on conventions and units.