Whether you’re just getting started with TabPFN or pushing it into production, this FAQ highlights practical answers to common questions about model limits, performance, reproducibility, and API best practices.Documentation Index
Fetch the complete documentation index at: https://docs.priorlabs.ai/llms.txt
Use this file to discover all available pages before exploring further.
What GPUs does TabPFN require at minimum?
What GPUs does TabPFN require at minimum?
At minimum, TabPFN requires an NVIDIA T4 GPU to run efficiently. For best performance, we recommend A100 or H100 GPUs. Any dedicated GPU supported by PyTorch is compatible, but some models may not have enough memory for larger datasets or perform slowly. CPU, MPS (Apple Silicon), and integrated GPUs are also supported, but are only suitable for small datasets.For larger datasets, fit + predict time can be dramatically reduced by parallelizing inference over several GPUs. To enable this, set the
device parameter of TabPFNClassifier and TabPFNRegressor.What are TabPFN's practical limits for context size and features?
What are TabPFN's practical limits for context size and features?
TabPFN is optimized for small- to medium-sized tabular contexts, since it performs in-context learning using transformer attention. See our Models for an overview of capabilities.
How are text features handled?
How are text features handled?
In the local package version text features are encoded as categoricals without considering their semantic meaning. Our API automatically detects text features and includes their semantic meaning into our prediction. The local package version encodes text as numerical categories and does not include semantic meaning.
How are date features handled?
How are date features handled?
In the local package version date features are encoded as categoricals without considering their semantic meaning. Our API automatically detects date features and creates an optimized embedding.
How reproducible are TabPFN results across runs and devices?
How reproducible are TabPFN results across runs and devices?
With a fixed seed and in the same environment TabPFN inference is deterministic. Across different hardware (CPU, GPU, MPS) configurations small differences are expected.
Does TabPFN handle missing values natively?
Does TabPFN handle missing values natively?
Yes. TabPFN’s estimators can handle missing values internally, including
pd.NA, without requiring manual imputation.How should I handle imbalanced datasets?
How should I handle imbalanced datasets?
TabPFN is inherently robust to imbalance. However, you can further optimize for specific objectives:
- Use
balance_probabilities=Trueto optimize for balanced accuracy or balanced loss. - Use
eval_metric="f1"(or other supported metrics) for specific precision-recall tradeoffs.
My run is 'slow' on thousands of rows - what controls the fit/predict trade-off?
My run is 'slow' on thousands of rows - what controls the fit/predict trade-off?
By default TabPFN performs most of its fitting inside the
predict() step (that’s how it simulates Bayesian posterior inference). That means latency scales roughly with the number of training rows. For fast inference we can create a tree- or small MLP based model that yields almost the same accuracy as TabPFN. Contact sales@priorlabs.ai to access this solution.How can I speed up predictions on large test sets?
How can I speed up predictions on large test sets?
Unless the fitted-model cache is enabled, the model is retrained each time
.predict() is called. It is much faster to make a prediction for all your test points in a single .predict() call rather than calling it repeatedly. If you run out of memory, split the test points into batches of 1,000 to 10,000 and call .predict() for each batch.You can also tune the memory_saving_mode and n_preprocessing_jobs parameters of TabPFNClassifier and TabPFNRegressor for additional speed improvements. See the code documentation for details.What is the fitted-model cache and when should I use it?
What is the fitted-model cache and when should I use it?
The fitted-model cache is an experimental feature that trains and stores the model during
.fit(), making subsequent .predict() calls fast by using a KV-Cache. Enable it by setting the fit_mode parameter of TabPFNClassifier or TabPFNRegressor to fit_with_cache.However, with this setting classification models consume approximately 6.1 KB of GPU memory and 48.8 KB of CPU memory per cell in the training dataset (regression models about 25% less), so it is currently only suitable for small training datasets. For larger datasets and CPU-based inference, we recommend the TabPFN-as-MLP/Tree output engine.What PyTorch version should I use?
What PyTorch version should I use?
TabPFN-3 requires PyTorch 2.5 or above; we recommend 2.8 or above for best performance.
What are the API rate limits?
What are the API rate limits?
Each API request consumes usage credits; the cost grows with the number of rows and columns in your dataset. You can check your current usage at ux.priorlabs.ai/api/usage, or from Python:See API metering for the full breakdown of token pools and per-model limits.
How do I know which TabPFN model version am I using in the TabPFN API Client?
How do I know which TabPFN model version am I using in the TabPFN API Client?
After making predictions with a TabPFN model using the client you can access model-level metadata, such as the exact model version used for inference. Each estimator exposes a .last_meta attribute containing this information: