Xpress Workbench is the technology component which provides an Integrated Development Environment (IDE) that allows Operations Research professionals, Data Scientists, and Solution Developers to rapidly and collaboratively develop analytic and optimization models, services, and Xpress Insight apps written in Python or Xpress Mosel. It also supports Xpress Mosel—the premier algebraic modelling and optimization language.
This release enables developers to debug Python/Insight 5 apps also locally in Xpress Workbench and provides several enhancements to users of Workbench on FICO Platform.
Python debugging in Workbench
Features only available on FICO Platform
For more information and detailed list of features and fixes, we recommend you read the Xpress release notes.
This release enables developers to debug Python/Insight 5 apps also locally in Xpress Workbench and provides several enhancements to users of Workbench on FICO Platform.
Python debugging in Workbench
- Python/Insight apps can now be conveniently debugged locally in Workbench before publishing to Xpress Insight.
Features only available on FICO Platform
- Availability of path-based routing: Workbench moves from host-based to path-based routing such that multiple components will now have the same URL domain name, but different paths. Therefore, the usual process of upgrading from earlier versions is not supported on DMP. Instead, users will need to create a new component and navigate to its backup archive page to restore the project backups from the original component.
- ‘debugpy’ library will be automatically included in Conda/Pip settings for local debugging if not already specified by the user.
- When logged into an Insight server, Workbench will compile Mosel using the version of mminsight suitable for that server (Insight v4 or v5). When downloading a zip, it will be built for the corresponding Insight version. Only if the users switch to an Insight server with a different version a full publish for the first time is required.
- Support for Mosel and Python apps to optionally configure dependencies via Conda, Pip, and R environment. Users can specify their own choice of R, Java and other libraries via files in their model_resources folder.
- File upload limit is raised from 100MB to 300MB.
For more information and detailed list of features and fixes, we recommend you read the Xpress release notes.
Updated 3 years ago
Version 3.0FICO
Community Manager
Joined July 21, 2025
FICO® Xpress Optimization Blog
Follow this blog board to get notified when there's new activity