# Start Here Biosimulant is a platform for discovering, running, and packaging biological simulations. Pick the surface that matches what you want to do today; all four paths share the same model formats, package refs, and simulation engine underneath. ## Choose your starting point
No install · about 5 minutes Run in the browser Browse a public lab, tune its inputs, run it, and inspect scientific results. Web quickstart → Python · about 10 minutes Build and run locally Install the open-source runtime and write a composable biological simulation. Library quickstart → API key · about 10 minutes Run durable managed jobs Submit a pinned Hub lab, track its lifecycle, and retrieve outputs and provenance. Developer API quickstart → Desktop and CLI Use the local workbench Use Studio Desktop as a graphical client around the same headless Python CLI available in terminals and CI. Desktop setup →
**Not sure?** Scientists exploring published models should start with the Web quickstart. Model authors should start with Python. Platform teams should use the Developer API, and local workflow or plugin authors should install Desktop. ## How it works Biosimulant is a unified framework for running biology models across any standard. The runtime does not care whether your model is an SBML pathway, a CellML cardiac cell, a NeuroML neuron, an ONNX inference module, or a few lines of Python. They all compose against the same contract. The contract is [`BioModule`](/references/library/biomodule-api). Subclass it. Declare your `inputs()` and `outputs()` as named ports with units. Implement `advance_window(start, end)` to move your state forward across one communication window. Return your outputs from `get_outputs()`. Any other module that follows the same contract can plug into yours. Connect `model_a.outputs.glucose` to `model_b.inputs.glucose` and the runtime handles unit checks, time alignment, and signal delivery. That is the entire framework. Once you've finished one quickstart, the [How-To Guides](/how-to) cover task-specific workflows (manifests, packaging, Labs Serve UI, BioWorld composition) and the [References](/references/library) cover the API and schemas in detail.