{"version":1,"type":"story","url":"https://digestai.news/story/nums-ai-launches-causilo-a-tabular-foundation-model-that-tops-tabarena","json":"https://digestai.news/story/nums-ai-launches-causilo-a-tabular-foundation-model-that-tops-tabarena.json","markdown":"https://digestai.news/story/nums-ai-launches-causilo-a-tabular-foundation-model-that-tops-tabarena.md","slug":"nums-ai-launches-causilo-a-tabular-foundation-model-that-tops-tabarena","headline":"Nums AI launches Causilo, a tabular foundation model that tops TabArena benchmarks","summary":"Nums AI released Causilo, a pretrained tabular foundation model for classification and regression, with a scikit‑learn‑style API, Apache‑2.0 code, and weights hosted on Hugging Face. It employs in‑context learning: the `fit` call stores training rows as context and predictions are made in a single forward pass without updating the pretrained weights.\n\nIn the TabArena benchmark covering 51 datasets, Causilo achieved the highest overall Elo (1794) among single models, surpassing Google Research’s TabFM and LG AI Research’s EXAONE Tabular. A re‑run that includes system ensembles placed it third of 88. On ScoringBench it also ranked first for CRPS, R² and RMSE across 101 regression datasets. Speed tests on an H100 GPU showed it fastest on both fit and predict, though TabPFN‑3 uses less memory.\n\nVersion 1.0.1 adds median and quantile outputs, and the model runs on CUDA or CPU with Python 3.10‑3.12 and PyTorch 2.13+. Commercial use requires a separate license; the released code and weights are for research only.","keyPoints":["Causilo 1.0.1 tops TabArena Elo (1794) among single tabular models, beating TabFM and EXAONE Tabular.","Uses in‑context learning: fit stores training rows as context, predicts queries in one forward pass without weight updates.","Runs fastest on an H100 GPU for both fit and predict; commercial deployment needs a separate license from Nums AI."],"whyItMatters":"Causilo demonstrates that large‑scale, pretrained tabular models can achieve state‑of‑the‑art performance with linear cost, opening doors for faster, more accurate data‑science pipelines and prompting broader adoption of foundation models beyond text and vision.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["Nums AI","Google Research","LG AI Research","Prior Labs","Xiaomi","Amazon"],"models":["Causilo","TabFM","EXAONE Tabular","TabPFN-3","Xiaomi-TabLDM","Mitra-v2"],"people":["Michal Sutter"]},"firstPublishedAt":"2026-09-16T04:59:47Z","updatedAt":"2026-09-16T04:59:47Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"MarkTechPost","title":"Nums AI Releases Causilo: A Tabular Foundation Model That Tops TabArena Among Single Models","url":"https://marktechpost.com/2026/09/15/nums-ai-releases-causilo-a-tabular-foundation-model-that-tops-tabarena-among-single-models","publishedAt":"2026-09-16T04:59:47Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Nums AI launches Causilo, a tabular foundation model that tops TabArena benchmarks\", 16 September 2026, https://digestai.news/story/nums-ai-launches-causilo-a-tabular-foundation-model-that-tops-tabarena","publisher":"Digest AI","title":"Nums AI launches Causilo, a tabular foundation model that tops TabArena benchmarks","datePublished":"2026-09-16T04:59:47Z","url":"https://digestai.news/story/nums-ai-launches-causilo-a-tabular-foundation-model-that-tops-tabarena"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}