Every prediction ships with its own uncertainty: conformal intervals and extrapolation flags. On the small datasets this platform exists for, strong R² with honest intervals is the realistic standard — and validation in the lab remains the final word.
Can I use my own models?
Not today. The platform trains its built-in ensemble per run; bringing custom models is on the roadmap. Your data and results, however, are always yours to export.
Who owns the outputs?
You do. You retain ownership of everything you upload and everything the platform generates from it — synthetic datasets, models, predictions, candidates and reports. See the Terms of use.
Institutional access
Institutional affiliation may unlock a higher-tier quota at registration. For institutional agreements or dedicated single-tenant capacity, contact the team.
Is my data used to train shared models?
No. Every model is trained on your data alone, for your account alone, and is never merged into a global or shared model. See the Privacy policy.