The data science revolution continues. TabPFN-3.5 is live, claiming SOTA beyond the vanilla IID small data setting and tackling data complexities that data scientists face everyday: ✅  grouped data ✅  temporal data ✅  strings in tables ✅  strong uncertainty calibration ✅  one model, whether it's classification or regression, 100 or 1.000.000 rows ✅  scaling to 20k features (benchmarked thoroughly to 6k) ✅  up to 6x faster inference with TabPFN-3.5-fast, plus FP-8 support on API Model report: storage.googleapis.com/prior… Try it today! 👉 Open-source repo: github.com/PriorLabs/tabpfn 👉 API: ux.priorlabs.ai/

Sep 15, 2026 · 3:30 PM UTC

4
30
191
12,124
Sort replies: Relevant Recent Liked
Replying to @FrankRHutter
I'd love a small worked example of the grouped split: train on some customers, test on entirely new ones. That would make the difference from an ordinary shuffled-row benchmark click for a lot of developers.
1
44
Good point, we built one now: docs.priorlabs.ai/cookbook/g…

Grouped data: TabPFN 3 vs 3.5 vs XGBoost - Prior Labs

Compare models on mice protein measurements, using grouped splits to evaluate predictions for unseen animals.

docs.priorlabs.ai
1
35
more replies
Replying to @FrankRHutter @pacoramon
Does it finally give uncertainty intervals over the class probabilities?
1
123
Replying to @FrankRHutter
You should probably explain what TabPFN actually is. Even your GitHub page doesn't explain it.
2
247
Replying to @FrankRHutter
I knew Prior Labs couldn‘t let other overtake them haha
31