soda-inria/tabicl

TabICLv2: A state-of-the-art tabular foundation model

63
/ 100
Established

This tool helps scientists, marketers, traders, and other professionals quickly get accurate predictions from their business data, without needing to spend time fine-tuning complex models. You provide a dataset with rows of observations and columns of features (like a spreadsheet), and it outputs predictions or classifications for new, unseen data. It's designed for anyone who needs reliable results from tabular data for tasks like predicting sales, identifying customer segments, or forecasting market trends.

603 stars. Actively maintained with 10 commits in the last 30 days.

Use this if you need highly accurate predictions or classifications from tabular data without extensive model configuration or hyperparameter tuning.

Not ideal if your dataset is extremely small (fewer than 300 samples) or if you need to deeply understand and interpret the inner workings of the model's decision-making process.

data-analysis business-intelligence predictive-modeling classification regression forecasting
No Package No Dependents
Maintenance 17 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

603

Forks

80

Language

Python

License

Last pushed

Mar 12, 2026

Commits (30d)

10

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/soda-inria/tabicl"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.