What SPM is
SPM is free, open-source desktop software for analyzing brain-imaging data sequences, including images from different cohorts and time series from the same subject. Its current release is designed for fMRI, PET, SPECT, EEG, and MEG analysis. SPM implements statistical methods for constructing and assessing spatially extended processes used to test hypotheses about functional imaging data. Listed capabilities include statistical tests, regression models, Bayesian analysis, and scripting. It imports NIfTI, Analyze, DICOM converted to NIfTI, OPM binary, TSV, JSON, FIF, and LVM formats. SPM is written in MATLAB and C and distributed under the GPL-2.0 license. The recommended installation is within MATLAB; GNU Octave and Standalone SPM are alternatives, with the standalone version running without a MATLAB license and including MATLAB Runtime requiring about 2.5 GB. Tutorials, datasets, and courses support learning. Community extensions include toolboxes, utilities, batch systems, and templates, though some may be incompatible. Development occurs on GitHub. SPM 25.01 is listed as the latest version. The SPM plan costs nothing.
Who it suits
SPM suits researchers analyzing fMRI, PET, SPECT, EEG, or MEG data with statistical, regression, Bayesian, and scripting tools. It is also relevant to users who can work with MATLAB, GNU Octave, or the standalone runtime.
Good to know, and what to watch
- Supports several neuroimaging modalities
- Includes statistical and Bayesian analysis
- Offers scripting support
- Imports many imaging formats
- Free and GPL-2.0 licensed
- Standalone runtime requires about 2.5 GB
- DICOM requires conversion to NIfTI
- Some community extensions may be incompatible
- Discussion-list replies from authors are not guaranteed
Plans
Every plan and priceSPM
Free plan
The verdict
SPM provides a broad neuroimaging analysis environment with multiple statistical approaches, scripting, and extensive format support. Choose the MATLAB, Octave, or standalone route based on your available runtime and workflow.
What the maker publishes
- Bayesian analysis
- Yes source
- Current release
- The software page lists SPM 25.01 as the latest version. source
- Data import formats
- NIfTI (.nii), Analyze (.hdr/.img), DICOM (converted to NIfTI), OPM binary/TSV/JSON, FIF and LVM source
- Deployment
- desktop source
- Development
- SPM development takes place on GitHub, where users can access development versions, report issues, request features and propose code changes. source
- Extensions
- Community-created extensions are listed as toolboxes, utilities, batch systems and templates, and the SPM developers disclaim responsibility for their usability and warn that some may be incompatible. source
- Founded
- 1991 source
- Free plan
- Yes source
- Implementation
- SPM is written in MATLAB and C. source
- Installation
- The recommended installation is within MATLAB; alternatives include GNU Octave and precompiled Standalone SPM without a MATLAB license. source
- Learning resources
- SPM provides online tutorials and datasets with instructions for analyzing them, and offers courses online and in person. source
- License
- SPM is open-source software available under the GPL-2.0 license. source
- Maker
- The SPM site identifies the Functional Imaging Laboratory at UCL Queen Square Institute of Neurology in London, UK, and says the software is made by its members and collaborators. source
- Platforms
- SPM works on Windows, macOS and Linux. source
- Purpose
- SPM analyzes brain imaging data sequences, including images from different cohorts and time series from the same subject. source
- Regression models
- Yes source
- Scripting support
- Yes source
- Standalone runtime
- Standalone SPM includes the MATLAB Runtime, which the documentation says requires about 2.5 GB. source
- Statistical methods
- SPM implements methods for constructing and assessing spatially extended statistical processes to test hypotheses about functional imaging data. source
- Statistical tests
- Yes source
- Support
- The SPM discussion list offers community discussion and help, but a response from an SPM author is not guaranteed because of resource limitations. source
- Supported modalities
- The current release is designed for fMRI, PET, SPECT, EEG and MEG analysis. source
Checked against the maker's pages on 10 Oct 2026.