NVIDIA opens Kumo Tabular at 215M parameters

NVIDIA open-sourced Kumo Tabular, a 28M-to-215M-parameter foundation model for zero-shot classification and regression that the company says ranks first across four tabular benchmarks. It was pretrained only on artificial tables and handles numerical and categorical columns, so MLPerf’s benchmark-quality boundary is the right companion lesson: run Kumo as a held-out baseline, then test calibration and distribution shift on your own data before replacing a tuned tree model.