NVIDIA open-sources Kumo Tabular foundation models for tabular data with permissive license
jure · x · 2026-09-29
NVIDIA has launched Kumo Tabular, a family of pretrained foundation models for tabular classification and regression, claiming a new Pareto frontier across the accuracy–inference-time tradeoff.
Key points:
- The models use labeled examples as context to predict outputs for new data, in-context learning style — no fine-tuning required
- Fully open: open weights, open-source software, and a permissive OpenMDW 1.1 license for commercial use
- Available on Hugging Face (nvidia/Kumo-Tabular); inference via the structured-data-models package with CUDA support
- Sample code shows building a TableTensor from a pandas DataFrame and running KumoTabular classification to get class probabilities
Related event: NVIDIA Open-Sources Kumo Tabular Foundation Models for Tabular Data(4 posts)→
More from Models
- Stealth Model 'silicon-2' Surfaces in Devin, Suspected to Be Anthropic's Next Model — kimmonismus · 2026-11-16
- Google hides Gemini model label in AI Mode, rolls Gemini 4 Carbon to Pro subscribers — gaganghotra_ · 2026-11-16
- Open-weight Karpo-Decision-31B tops Decision Index at 73.9, one forward pass per answer — CodeByPoonam · 2026-11-16
- Nathan Lambert calls out Ethan Mollick for 'selectively ignoring' Kimi in model comparisons — natolambert · 2026-11-16
- Eval Firm Trajectory Told 40 Agents "NEVER Stop" in Harness Without Telling Epoch — peterwildeford · 2026-11-16
- Unsloth Q8 crushes default Q3 in a self-hosted test over 331 financial PDFs — No_Cartographer3953 · 2026-11-16