Our toolchains, accessible using a variety of programming languages, provide an easy way to create compute tasks that is directly integrated into your workflow.

See below for a short introduction to our Sclblpy package. For a more extensive, step-by-step walkthrough from Python to online app, see our getting-started tutorial: Scailable 101: Getting started.

sclblpy Python package

The sclblpy Python package can be used to upload fitted models directly to Scailable. For the full documentation please see

The process is simple:

  1. First, install the python package directly from github using pip:
    pip install sclblpy
  2. Next, fit your favorite model using the tools you know and directly upload it to Scailable. For example:
    # Neccesary imports:
    import sclblpy as sp
    from sklearn import svm
    from sklearn import datasets
    # Start fitting a simple model:
    clf = svm.SVC()
    X, y = datasets.load_iris(return_X_y=True), y)
    # Create an example feature vector (required):
    row = X[130, :]
    # Create documentation (optional, but useful):
    docs = {}
    docs['name'] = "My first fitted model"
    docs['documentation'] = "Any documentation you would like to provide."
    # Upload the model:
    sp.upload(clf, row, docs=docs)
  3. Finally, you will be notified by email when your model has been converted to wasm and can be consumed as a REST endpiont. Your endpoint will be added to your list of endpoints and you can take it from there!

For the full package documentation please see The current beta release supports a large range of models from sklearn, statsmodels, and xgboost.

If you need more functionality or a wider range of models, please let us know at

sclblR [R] package

Scailable also allows you to upload models directly from R. The sclblR package is however not available in the current beta.

If you want to try out sclblR, let us know at

Created in the Netherlands with ♥. Scailable is committed to responsible AI. Questions? Contact us.