Skip to content
SPECIFICTION
Get listed

REaLTabFormer vs Synthehol Dataset

REaLTabFormer

github.com

Synthehol Dataset

dataset.synthehol.ai

Side by side

FigureREaLTabFormerSynthehol Dataset
Price fromFree$332.50/mo
Free planYesYes
Free trial—Yes
Open source——
Web—Yes
WindowsYes—
macOSYes—
LinuxYes—
API—Yes
Self-hostedYesYes
Plans published17
Spec score · AI Synthetic Data Generators8.5 (#4)8.7 (#3)
PurposeREaLTabFormer is a framework for generating synthetic tabular and relational data with transformer models.Synthehol Dataset generates statistically accurate, privacy-preserving synthetic records from a sample dataset.
SupportFor vulnerability reports, the policy lists [email protected] and requests details that help reproduce and assess the issue.Support ranges from basic support on Free to email support on Starter and priority plus Slack support on Professional.

Plans

REaLTabFormer

  • REaLTabFormerFree

REaLTabFormer pricing →

Synthehol Dataset

  • FreeFree
  • FreeFree
  • Starter$399 a month
  • Professional$2,999 a month
  • Starter$3,990 a year
  • Professional$29,990 a year
  • EnterprisePriced on request

Synthehol Dataset pricing →

In their own words

REaLTabFormer

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.

Visit github.com

Synthehol Dataset

Synthehol Dataset generates synthetic records from a sample dataset, with the stated aim of protecting privacy while reflecting statistical patterns. The workflow starts with uploading a sample, then analyzes distributions and correlations before generating rows. Each dataset is measured for privacy, fidelity, utility and similarity. The site says the platform can produce 10 million rows in about 12 seconds. Access is available through web, API and self-hosted options; the platform page advertises REST API access and batch generation. CSV and JSON exports are included on Free and Starter, while Professional also lists Parquet and Avro. Synthehol says customer data is not sold, shared or used except to provide its services, and describes the product as built for HIPAA, GDPR, SOC 2 and ISO 27001. The maker presents it for ML training pipelines, QA fixture generation and analytics workloads, across use cases including healthcare, finance, pharmaceuticals, insurance, retail and load testing. Free includes 1,000 monthly credits and up to 50,000 rows per month; paid plans start at $399 a month. Enterprise pricing is custom and includes an on-premise option.

Visit dataset.synthehol.ai