Run SAS Code Online

No install, no sales call, and no card to start. Sign in and a JupyterLab workspace opens in your browser with the Jenner engine already in it: paste a program, run it, read the log.

The short answer

Create an organization and a JupyterLab workspace opens in your browser with Jenner installed, so a .sas program can be running within a few minutes of arriving. Every new organization starts with trial credits and no card, which is enough to run real programs before deciding anything. Jenner is an independent implementation of the SAS language, so the file you already have is the file that runs — nothing is translated or rewritten on the way in.

What the first ten minutes look like

Upload a .sas file and the data it reads, or paste the program straight into a notebook cell. Run it and the log comes back in the shape you expect — notes, warnings and errors attached to the statements that produced them. When something is unsupported the log says which statement it was, which is the behaviour you want from anything you are evaluating.

What you can run in the browser

The language your programs are already written in, executing on a machine you did not have to provision. This is not a converter: a DATA step stays a DATA step and PROC SQL stays PROC SQL.

A workspace, not a snippet box

You get a full JupyterLab session — file browser, notebooks, a terminal, and somewhere to keep the data your program reads. A real program with several files and a LIBNAME behaves like a real program, which is the part single-file playgrounds cannot do.

DATA steps, PROC SQL, and macros

The core of the language runs as written: DATA step logic, arrays, formats and informats, the macro facility, and PROC SQL. Code written against SAS 9.4 syntax does not need a porting pass before it will execute.

100 procedures

Regression and linear modeling, survival analysis, multivariate methods and clustering, time series, quality control, summary and reporting. Every procedure has a published reference page listing the statements and options it supports, so you can look up the ones your program needs before you upload anything.

Your SAS7BDAT files, read in place

Jenner reads and writes SAS7BDAT and XPT directly, alongside CSV and Parquet. Upload the dataset you already have, point a LIBNAME at it, and read it the way your program does now — no export step, and no third format in the middle to lose a label in.

How to tell whether it runs your code

Run your own programs and read the log. That is the only test that settles it, and a browser workspace is the fastest way to get to it — coverage is not the whole SAS language, so a demo on someone else's example proves very little about yours. Start with the program that matters most, compare its output against numbers you already trust, and treat whatever the log flags as the real answer.

Frequently Asked Questions

Create an organization, open a workspace, and you are in a JupyterLab session in your browser with the Jenner engine already installed. Upload the .sas file or paste it into a cell, then run it. Nothing is installed on your machine, and the program is executed as written rather than converted first.

Yes, to start. Every new organization begins with trial credits and no card, which is enough to run real programs and decide whether this answers your question. What there is not is a permanently free workspace tier: once the credits are gone, workspace time is billed by the hour, with no per-seat licence and no subscription that resets. Two neighbouring products differ — the desktop app downloads free with a row cap, and the HTTP API has an anonymous tier open to everyone.

SAS OnDemand for Academics is SAS Institute's own browser-based environment, and for a lot of readers it is the right answer, because it runs real SAS. Who may use it, and on what conditions, are SAS's terms to state rather than ours — read their current terms before you build anything on it. People usually arrive here when those terms do not fit the work they need to do.

Altair SLC also runs SAS-language programs without SAS installed, and it has been doing that far longer than we have. Where someone else's edge case has already been hit and fixed, that track record is worth real money, so compare both against your own code rather than against a feature list. Siemens now sells it as Rapidminer SLC, so it is worth searching under both names.

Yes, directly. SAS7BDAT is read and written in place: upload the dataset, point a LIBNAME at the folder, and read it the way your program reads it now. XPT, CSV and Parquet work the same way, so there is no export step and no intermediate format for a variable label or a format to go missing in.

Not through a library of ours — there is no Python or R package to install. What there is: the engine runs as a command-line program, so either language calls it the way it calls any other command-line tool, and it also runs over HTTP, so a program can be sent and its output read back from a script. That is usually how it ends up inside a pipeline rather than a notebook.

It is a clean-room implementation of the SAS 9.4 language, written in Rust. Clean-room means it was built from the published behaviour of the language, not from SAS software: nothing of SAS's is embedded, wrapped or called, which is why no SAS licence is involved at either end.

Run them and read the log, which is the whole reason the workspace opens without a call. Coverage is not the whole SAS language, so some programs will hit something that is not supported — when that happens the log names the statement rather than quietly returning a number that looks plausible. Look up the procedures you depend on in the reference before you commit to anything.

On our infrastructure. A hosted workspace means the files you upload sit on machines we operate, which is the trade you make for not installing anything. If that does not suit the data, the same engine ships as a native application for Mac, Windows and Linux and as a command-line binary you run on your own hardware; the programs do not change when you move between them.

No, and the difference matters. Several results for this search convert code instead of running it, which hands you a new program to validate from scratch — a different job with different failure modes. Here the .sas file is executed as it stands, so what you tested is what you keep.

Run your first program in the browser

Open a workspace and paste in the program you most want to check. No install, and no card to start.

Start Free Trial