MathWorks Launches MATLAB Copilot and Redesigned Desktop in 2025a Release
MathWorks has shipped its 2025a release of MATLAB and Simulink, marking one of the most significant updates to the platform in years. The centerpiece is MATLAB Copilot, a generative AI assistant embedded directly into the MATLAB environment, alongside a fully redesigned desktop interface and a broad set of enhancements across simulation toolboxes.
MATLAB Copilot: AI-Assisted Engineering Workflows
MATLAB Copilot is an AI assistant purpose-built for technical computing. Unlike general-purpose coding assistants, it is trained on MathWorks documentation, code examples, and domain-specific engineering knowledge. Key capabilities include:
- Context-aware code generation: Describe a task in natural language and Copilot generates syntactically correct MATLAB code, including calls to specialized toolboxes.
- Inline explanations and debugging: Copilot can explain unfamiliar code snippets, clarify error messages, and suggest fixes—reducing the time engineers spend deciphering legacy scripts.
- Automated test generation: Integration with MATLAB Test allows Copilot to generate unit tests from existing functions, improving verification coverage with minimal manual effort.
- Auto-completion and prediction: Real-time code suggestions appear as engineers type, accelerating development of simulation scripts and data analysis pipelines.
For simulation engineers, this means faster construction of parametric studies, automated generation of model initialization scripts, and lower barriers to adopting new toolboxes.
Redesigned Desktop Environment
The 2025a release introduces a modernized MATLAB desktop with customizable sidebars that surface tools such as the Debugger, Source Control, Test Browser, and Code Issues panel without requiring separate windows. A new Figure Container consolidates plots and visualization tools into a single, tileable workspace—particularly useful when monitoring multiple simulation outputs simultaneously.
Light and dark themes are now supported across the desktop, figures, and apps, and a new Desktop Search bar provides instant access to documentation, examples, and MATLAB Answers from within the IDE.
Simulink Enhancements for System Simulation
On the Simulink side, the 2025a release delivers several productivity improvements relevant to control system and physical modeling workflows:
- Python Code Block: Engineers can now embed Python code directly inside Simulink models, enabling seamless integration of Python-based machine learning models or data processing routines into hardware-in-the-loop and model-in-the-loop simulations.
- Local Solvers for Model References: Referenced models can now use independent variable-step or fixed-step solvers, improving both simulation speed and numerical accuracy for multi-rate systems.
- Fast Restart for Rapid Accelerator: Repeated rapid accelerator simulations—common in Monte Carlo and design-of-experiments workflows—execute faster thanks to reduced recompilation overhead.
- New Simulink Scope: A redesigned scope UI with multithreaded rendering and direct connectivity to the Simulation Data Inspector streamlines signal monitoring during long simulation runs.
Toolbox Highlights
Several toolboxes received targeted updates:
| Toolbox | Key Update |
|---|---|
| RoadRunner / RoadRunner Scenario | New programmatic API for creating road scenes and automotive scenarios |
| Sensor Fusion & Tracking | Tracking Data Importer app; simplified multi-object tracker tuning |
| Radar Toolbox | Parallelizable workflows for bistatic and multistatic radar simulation |
| MATLAB Coder | New app UI; automatic parallelization for generated C/C++ code |
| Phased Array System Toolbox | Reconfigurable Intelligent Surface (RIS) modeling support |
Python and Source Control Integration
The 2025a release deepens Python interoperability: MATLAB string arrays now automatically convert to Python lists or NumPy string arrays, and Python environments can be managed directly from the MATLAB desktop. Live scripts can be saved in a plain-text .mlx format for improved compatibility with Git and other version control systems—a long-requested feature for teams managing simulation models in collaborative repositories.
Availability
MATLAB and Simulink 2025a is available now through MathWorks' standard licensing channels. The 2025b release, focused on stability and refinements, followed later in the year. Full release notes and migration guides are available on the MathWorks website.
For simulation engineers evaluating AI-assisted workflows, the MATLAB Copilot represents a meaningful step toward reducing the scripting overhead that has historically been a barrier to rapid simulation iteration.