Tabular data

Kumo Tabular

A pretrained model for tables: classification and regression from in-context examples, with no task-specific training. You supply labeled rows and the rows that need a forecast and get class probabilities.

Developer
NVIDIA, USA
First release
Sep 2026
Latest release
Sep 2026
Sizes
size not stated on the card
License
Commercial use allowedOpenMDW 1.1
Running
On your own serverNeeds a GPU
Industries
Finance, Retail and marketplaces, Manufacturing and logistics
Deploy on your server Hugging Face GitHub

What it does

  • scoring customers and requests
  • forecasting from a table of sales or orders
  • detecting churn from a customer table

Where it is used

finance and scoringretail and salesanalytics

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. Kumo Tabular

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can Kumo Tabular be used in a commercial project?

Yes. License: OpenMDW 1.1. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does Kumo Tabular need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Without a GPU the model is not practical. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does Kumo Tabular support Russian?

Language does not matter for this model: it does not work with text.

Where can I download Kumo Tabular and what does it cost?

The Kumo Tabular weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Tabular dataTabPFNPrior Labs (University of Freiburg) · GermanyCommercial use with conditions

A ready-made model for tables: it takes example rows and immediately predicts for new ones, without lengthy training or tuning. Only v2 is free for business; newer versions are non-commercial.

DetailsTabular dataTabICLInria (Soda team) · FranceCommercial use allowed

An open tabular model from the creators of scikit-learn: classifies and predicts from examples without training and handles tables of up to hundreds of thousands of rows. The license allows business use.

DetailsTabular dataXiaomi-TabLDMXiaomi · ChinaCommercial use with conditions

A small tabular model: it takes labeled rows as context and predicts values for new rows in one pass, with no task-specific training. Two checkpoints: classification and regression.

Details

Source: huggingface.co/nvidia/Kumo-Tabular. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.