Explore & download data

Synthetic bureau-style attributes for credit risk modeling.

Synthetic loan applications with risk drivers, bureau-style fields, and offer outcomes.

Wide tabular features for subscription churn modeling: usage, billing, support, and product mix.

Subscription and usage features with churn label for binary classification.

Synthetic customer features for clustering and RFM-style segmentation. Wide table with 18+ columns for big-data analytics.

Large wide table with 24+ columns for big-data pipelines: user, session, product, and behavioral features. Suited for ML feature stores and analytics.

Bipartite-style edges for graph ML on shared devices and accounts.

Per-session engagement, monetization, and technical KPIs for live-ops analytics.

Synthetic housing features and sale prices for regression and feature importance. Wide table with 18+ columns for big-data modeling.

Wide employee-level features for turnover and retention models.

Synthetic benchmark runs across models—latency, token usage, accuracy, cost, and judge scores for reproducible LLM evaluation pipelines and A/B routing research.

Synthetic loan applications with approval flag for classification and fairness analysis.

Product attributes and stock levels for demand forecasting and replenishment.

Synthetic applicant profiles with scores and outcomes for fair ML benchmarking.