TabICL 2.2.0 ships with pretraining and prior features, plus fixes for time-series and fine-tuning
RichmanRonald · x · 2026-09-03
TabICL, the tabular foundation model, released version 2.2.0 with new pretraining and prior features, improved MPS/XPU inference, and a batch of bug fixes covering time-series handling, fine-tuning, threading, memory limits, and numpy object arrays. The project is open source on GitHub.
More from Research
- New arXiv paper quantifies CoT necessity via opaque serial depth amid OpenAI architecture rumors — ArthurConmy · 2026-09-04
- Tabular Deep Learning for Trading: Cross-Regime Bayesian Optimisation for Equity Signals — PtrPomorski · 2026-09-04
- Nature Human Behaviour: short funding cycles and productivity metrics stifle creative science — S_OhEigeartaigh · 2026-09-03
- A huge 176-digit integer is claimed to divide RSA-260, unverified — CatAstro_Piyush · 2026-09-03
- Kimi K3 Draft Collection released: EAGLE-3, DFlash2 and DSpark draft models trained on GB200 — hongyangzh · 2026-09-03
- Nature Biotech paper: reference materials as common calibrator to make multiomics data AI-ready — kshameer · 2026-09-03