Google Releases TabFM, a 400M-Parameter Foundation Model for Tabular Data

Google introduced TabFM, a 400M-parameter tabular foundation model trained purely on synthetic data from structural causal models, achieving zero-shot top performance across 51 TabArena benchmarks, beating tuned AutoML baselines.

2026-09-30 ~ 2026-10-01 · 2 related posts

• Google's TabFM: 400M-param tabular foundation model beats tuned AutoML zero-shot — google · 2026-09-30
• TabFM: 400M-parameter tabular foundation model beats tuned AutoML zero-shot on all 51 TabArena datasets — rickasaurus · 2026-10-01