Models

The following table summarizes the key differences between the major TabPFN releases. The “Max Rows” and “Max Features” represent the recommended maximum dataset sizes the models were designed and evaluated for. * More information about our license can be found here.

​

TabPFN-2.5

• Publication: Detailed documentation and release notes will be available soon.
• License: Model checkpoint under: TABPFN-2.5 License v1.0, code under Prior Labs License (Open source, commercial use with attribution)

TabPFN-2.5 is the next generation of the tabular foundation model, designed to scale to significantly larger datasets. It expands the in-context learning capabilities of TabPFN to additional tabular problems. Key Features:

• Scales to 50,000 samples and 2,000 features.
• Continues to support mixed data types and requires no preprocessing.

TabPFN-2.5 Technical Report

Read the full TabPFN-2.5 technical report.

​

TabPFNv2

• Publication: “Accurate predictions on small data with a tabular foundation model” (Nature, 2025)
• License: Prior Labs License (Open source, commercial use with attribution)

TabPFN-2 is the model published in Nature and represents a significant upgrade over v1. It was re-engineered to handle real-world, “messy” data. It remains a powerful and accessible foundation model for small- to medium-sized datasets. Key Features:

• Handles mixed data types (numerical and categorical) and missing values automatically.
• Scales to 10,000 samples and 500 features.
• Delivers SOTA performance, often outperforming tuned Gradient-Boosted Trees (GBTs) and AutoML systems on datasets within its size limit.

​

TabPFN (v1)

• Publication: “TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second” (ICLR 2023)
• License: Apache 2.0

This is the original TabPFN model presented at ICLR 2023. It introduced the concept of a Prior-data Fitted Network and demonstrated that a transformer pre-trained on synthetic data could achieve strong results on small tabular datasets within a second. Key Limitations:

• Designed only for numerical features.
• Strictly limited to datasets with 1,000 samples and 100 features.

​

TabPFN-2.5 Model License

What does the non-commercial license mean for enterprise use?

Enterprises can download and experiment with the model internally without cost. This includes benchmarking, testing on internal datasets, exploring capabilities, and understanding fit for use cases. Any competitive benchmarking (i.e. for procurement), testing in deployment in production environments, use in workflows that support business decisions, client projects, or revenue generating activities requires a separate commercial license or API agreement.

Can enterprises use TabPFN-2.5 for internal testing?

Yes. Enterprises may run preliminary internal assessments on their own data for research and exploration. This includes testing on internal datasets, exploring capabilities and understanding fit for use-cases.

What is not allowed in enterprise testing?

You cannot use the model or its outputs for any activity that influences business decisions, evaluation of vendors, procurement decisions, or commercial workflows. This includes proof of value that affects internal budgeting or vendor selection. In those cases, a commercial license is required. Any competitive benchmarking (i.e. for procurement), testing in deployment in production environments, use in workflows that support business decisions, client projects, or revenue generating activities requires a separate commercial license or API agreement.

Can we use results from testing to inform product features or deployment strategy?

Not without a commercial license. Any evaluation that influences commercial decisions or internal production planning is considered commercial use.

Can we evaluate TabPFN-2.5 for integration into our product?

You can run small scale internal experiments. If you intend to integrate TabPFN-2.5 into a product or platform, the API or a commercial license is required.

If we want to move from testing to production, what is required?

You will need a commercial license or API agreement. Contact sales@priorlabs.ai.

Do non-commercial rules also apply to derivative models or fine tuned versions?

Yes. The same restrictions apply to all derivatives. Fine tuned, modified, or otherwise adapted models created from TabPFN-2.5 remain non-commercial unless you secure a commercial license.

What about outputs generated by the model?

Outputs are yours, but they can only be used for non-commercial evaluation and research. Using them in production or for any business decisions requires a commercial license.

If we train a model from scratch without the weights, is that covered by this license?

If you do not use any TabPFN-2.5 model weights, checkpoints, or other licensed model elements, then the non-commercial license does not apply to that separately developed model. However, if you use TabPFN-2.5 or its outputs to train, guide, fine tune, or distill another model for commercial use, that is not permitted under this license.

Can we use TabPFN-2.5 to train a commercial model indirectly?

No. Using TabPFN-2.5 or its outputs to train, fine tune, distill, or otherwise improve a commercial model is not allowed without a commercial agreement.

Where should we go if we want to deploy TabPFN in production or test it?

Use our API or obtain a commercial license. The API is the simplest path for production trials and product integration.

What about the previous TabPFNv2 release?

If you want to continue using the earlier TabPFNv2 release, it remains available under its original license. This non-commercial license applies only to TabPFN-2.5. The terms for prior versions do not change.

Who do we contact for commercial licensing?

Please reach out to our commercial team at sales@priorlabs.ai.