- Home
- Synthetic Data Generation Software
- SynthCity
SynthCity
What it is
SynthCity is a free Python library for generating and evaluating synthetic data. Its generators cover tabular data, time series, survival analysis, and images, with approaches such as GANs, VAEs, normalizing flows, Bayesian networks, random forests, and LLM-based generators. A plugin architecture is designed to support extensions. Evaluation metrics address data correctness, quality, and privacy, while its Benchmarks API can assess generators with selected metrics and repeated runs. Privacy-focused methods include DECAF, DP-GAN, AdsGAN, PATEGAN, and PrivBayes; DECAF also supports inference-time debiasing based on user-defined fairness requirements. Demonstrations use scikit-learn and lifelines datasets. The library can be installed from PyPI or source, and serialization utilities save and load models from files or buffers. SynthCity is self-hosted and Linux-focused, and the repository identifies its license as Apache-2.0. Missing data must be imputed before use; HyperImpute is suggested for that task. The listed image-generator architectures are described as not state of the art, with extension suggested for other architectures.
For and against
For
- Supports multiple synthetic data modalities.
- Includes metrics for correctness, quality, and privacy.
- Benchmarking supports selected metrics and repeated runs.
- Model serialization supports files and buffers.
- Identifies its license as Apache-2.0.
Against
- Missing data must be imputed first.
- Listed image architectures are not state of the art.
- Linux-focused and self-hosted.
The verdict
SynthCity offers extensible generators and evaluation tools across several data modalities. Users need to impute missing data first, and the listed image architectures have a stated limitation.
Who it suits
SynthCity suits Python users building or evaluating synthetic data generators, including those working with tabular, time-series, survival, or image data. It may also suit users assessing privacy or fairness-related methods.
Plans and prices
SynthCity
Open-source Python library · install from PyPI or source
Read 3 Oct 2026 · github.com
On the record
- Benchmarking
- The Benchmarks API can evaluate generators using selected metrics and repeated runs.github.com · read 3 Oct 2026
- Data modalities
- The repository describes generators for tabular data, time series, survival analysis, and images.github.com · read 3 Oct 2026
- Evaluation
- It includes evaluation metrics for data correctness, quality, and privacy.github.com · read 3 Oct 2026
- Extensible architecture
- The library has a plugin architecture designed to be extended.github.com · read 3 Oct 2026
- Fairness and privacy
- Privacy-focused methods include DECAF, DP-GAN, AdsGAN, PATEGAN, and PrivBayes, and DECAF supports inference-time debiasing using user-defined fairness requirements.github.com · read 3 Oct 2026
- Generator types
- The listed models include GANs, VAEs, normalizing flows, Bayesian networks, random forests, and LLM-based generators.github.com · read 3 Oct 2026
- Image model limit
- The README says its image generator architectures are not state of the art and suggests extending the image generator/discriminator architecture method for other architectures.github.com · read 3 Oct 2026
- Installation
- The library can be installed with pip from PyPI or installed from source.github.com · read 3 Oct 2026
- Integration
- The README demonstrates using SynthCity with scikit-learn datasets and lifelines survival datasets.github.com · read 3 Oct 2026
- License
- The repository identifies its license as Apache-2.0.github.com · read 3 Oct 2026
- Missing data limit
- SynthCity does not handle missing data, which must be imputed first; the README points to HyperImpute for this purpose.github.com · read 3 Oct 2026
- Model serialization
- Models can be serialized and loaded from buffers or files using the library's serialization utilities.github.com · read 3 Oct 2026
- Purpose
- SynthCity is a library for generating and evaluating synthetic tabular data.github.com · read 3 Oct 2026
- Support and contributions
- The project tracks public bugs through GitHub Issues and welcomes pull requests.github.com · read 3 Oct 2026
