FrontierSeptember 16, 2026via MarkTechPost
Prior Labs Releases TabPFN-3.5: A Tabular Foundation Model That Beats the Winning Otto Kaggle Solution With Default Settings
Why it matters
A new tabular foundation model trained only on synthetic data outperforms the winning solution from a major ML competition with default settings. This signals a shift in how specialized domains (tabular ML) are approaching foundation models and suggests synthetic pretraining may be sufficient for competitive performance.
Key signals
- Prior Labs released TabPFN-3.5
- Model beats Otto's Kaggle competition winning solution
- Achieves competitive performance with default settings (no tuning required)
- Trained exclusively on synthetic data
- Tabular foundation model approach
- Published September 2026
The hook
TabPFN-3.5 beats Kaggle's best without tuning—synthetic data just entered the foundation model era.
Prior Labs released TabPFN-3.5, a tabular foundation model pretrained only on synthetic data that beats Otto's winning solution.
The post Prior Labs Releases TabPFN-3.5: A Tabular Foundation Model That Beats the Winning Otto Kaggle Solution With Default Settings appeared first on MarkTechPost.