TabPFN
TabPFNv2.5CurrentReleased January 1, 2025
Context--
Training CutoffTabPFN-2.5 model report era
CategoryTabular Foundation Model
About TabPFN
TabPFN is Prior Labs’ tabular foundation model for classification and regression on structured data. It returns predictions in one forward pass and reduces the need for manual model tuning.
Capabilities
classificationregressionanomaly detectiontime series forecastingsynthetic data
Input Modalities
tabular-datacsvdataframe
Output Modalities
predictionclassificationregression
Technical Details
- API Identifier
- priorlabs/tabpfn-2.5
- Category
- Tabular Foundation Model
Tags
tabular-aifoundation-modelclassificationregressionstructured-data
Run this model yourself
Local and cloud setup for TabPFN, with exact configurations, sources and clear evidence limits.
No reviewed setup guide is available for this model yet. Explore configurations with documented support in the hardware finder.
Benchmarks
Performance scores for TabPFN across standard benchmarks.
TabArena vs tuned tree-based modelsofficial-prior-labs · May 2026
100
Maximum supported data pointsofficial-prior-labs · May 2026
50000
Maximum supported featuresofficial-prior-labs · May 2026
2000
Pricing and access
Prior Labs offers a hosted API and a non-commercial open-source package; public per-token pricing was not found in reviewed sources.
Prior Labsstartup
Tabular foundation models for structured data
1 ModelsFounded 2024Freiburg, Germany
View full profile