REaLTabFormer vs Synth Studio

REaLTabFormer is a free, MIT-licensed framework for generating synthetic tabular and relational data. Its relational model uses a sequence-to-sequence approach, while its model for independent tabular observations uses GPT-2. Examples use pandas DataFrames as input; relational generation requires matching join-key columns in the parent and child tables. The documented workflow fits a model, saves it locally and samples synthetic data. For non-relational training, it stops when the synthetic distribution is close to the real data distribution. Observation validators can filter invalid samples, including through a GeoValidator example. The paper describes target masking to prevent data copying and the Qδ statistic with statistical bootstrapping to detect overfitting. Installation uses `pip install realtabformer`; the current PyPI package requires Python 3.8 or newer. PyPI classifies it as operating-system independent, and Linux, macOS, Windows and self-hosted platforms are listed. The project describes use in projects or research and asks users to cite its research paper when using it. For vulnerability reports, its policy directs users to email rather than post publicly and says a response should arrive within 48 hours.

Synth Studio generates synthetic tabular datasets with mathematical privacy guarantees. It accepts CSV and JSON files and detects their schemas automatically. Generation options include CTGAN, TVAE, GaussianCopula and schema-based methods. Differential privacy settings can be configured with epsilon and delta, using RDP accounting. Quality tools assess statistical similarity, machine-learning utility and privacy leakage, while automatic PII detection identifies fields such as names, emails and SSNs before generation. The product provides compliance reporting for HIPAA, GDPR, CCPA and SOC-2, alongside audit logs and model cards. Synth Studio is open source and can be self-hosted or used through its cloud offering. Local installation documentation lists Windows 10+, macOS 10.15+ and Linux. The free plan is free for up to 1 million rows per generation under an MIT License. REST API integration is available, but direct API access currently requires a session cookie or trusted proxy headers; API keys are upcoming. Documented API limits include 1,000 authenticated requests per minute and five concurrent generation jobs.