YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
QuantScenarioBench
QuantScenarioBench is an open-source, JAX-native Python library for generating reproducible stochastic market scenarios. It provides a common state-space interface across models — Black-Scholes, Heston, and Rough Bergomi in v1 — so researchers can simulate, compare, and export price paths with a single simulate() call, with bit-identical reproducibility on a given backend.
About this organization
This organization hosts the official benchmark sample datasets for QuantScenarioBench.
These datasets are lightweight demonstration samples, not comprehensive research datasets. Each one uses a single fixed configuration (10,000 paths, daily steps over a 1-year horizon, seed 42) so they load quickly and support direct, reproducible comparison across models. They are not intended for research-scale experiments.
For custom datasets — arbitrary horizons, time grids, model parameters, or numbers of paths — use the QuantScenarioBench library directly and export with export_parquet() or publish_to_hub(). See the GitHub repository for installation and full documentation.
Available datasets
| Model | Dataset |
|---|---|
| Black-Scholes | QuantScenarioBench/qsb-black-scholes |
| Heston | QuantScenarioBench/qsb-heston |
| Rough Bergomi | QuantScenarioBench/qsb-rough-bergomi |
All three datasets share the same 12-column schema, time grid, seed, and initial spot for direct cross-model comparison.