What Quick ML is

Quick ML is a pocket data studio for iPhone and iPad. You import a table of data, explore and clean it, train a machine learning model, and use that model to predict new values. All of it happens on the device in your hand. Nothing you import or train is sent to a server, because there is no server.

Full disclosure before we go any further: I made it. This guide is a walkthrough of my own app, so read it as one person showing you round their kitchen rather than an impartial review. I have tried to keep it honest about what the app does not do, as well as what it does.

The loop

Every project in Quick ML follows the same shape, and the app lays it out for you as a workflow of three stages.

• Data is where the table comes from and what you do to it. Files of any size, web pages, APIs, Kaggle datasets and PostgreSQL databases all come in here. Cleaning is recorded as a recipe of steps, which we will get to.
• Train is where you pick what to predict and which algorithm to use. There are nineteen algorithms across regression, classification, clustering and image classification.
• Predict gives you a form generated from your model, so you can type in new values and see what the model thinks. It will also run a whole file through the model in one go.

Beside those, a Dashboard composes live charts and stat tiles from your data, and a PDF Report writes the whole project up in one document.

Who it is for

If you have ever opened a spreadsheet and wondered whether the numbers in it could predict something, Quick ML is for you. You do not need to know Python. The app uses plain English throughout, and it has a built-in glossary that explains every statistical term it uses, from R squared to SMOTE.

If you do know Python, there is a treat for you too. Every project can be turned into a Jupyter notebook that does the same cleaning and training in pandas and scikit-learn, one explained cell per step.

What it costs

The free version is a real, working app rather than a demo. You can import files of any size, clean without limits, train the classic algorithm for each task, evaluate honestly, predict, and produce the PDF report. Quick ML Pro is a single purchase that unlocks everything else. There is no subscription and no account to create. The last chapter of this guide lists exactly which features sit on which side of the line.

What you need

An iPhone or iPad running iOS 26 or later. Projects sync between your own devices through iCloud, so you can import on the iPad and predict on the phone.

Quick ML on the App Store

Free to download for iPhone and iPad. One purchase unlocks Pro.

Open

How this guide is organised

The chapters follow the loop. We start with a project and a file, then work through the other ways of getting data in, then cleaning, then each kind of model, then predicting, dashboards and exports. The image classifier and the time-ordered chart get their own chapters at the end. If you only want to see something work, the second chapter takes you from nothing to a trained model in about five minutes using the sample data built into the app.