{"url":"/method/tabpfn","slug":"tabpfn","name":"TABPFN","full_name":"tabular data Prior-data Fitted Network","full_name_withheld":false,"description_markdown":"We present TabPFN, a trained Transformer that can do supervised classification for small tabular datasets in less than a second, needs no hyperparameter tuning and is competitive with state-of-the-art classification methods. TabPFN is fully entailed in the weights of our network, which accepts training and test samples as a set-valued input and yields predictions for the entire test set in a single forward pass. TabPFN is a Prior-Data Fitted Network (PFN) and is trained offline once, to approximate Bayesian inference on synthetic datasets drawn from our prior. This prior incorporates ideas from causal reasoning: It entails a large space of structural causal models with a preference for simple structures. On the 18 datasets in the OpenML-CC18 suite that contain up to 1 000 training data points, up to 100 purely numerical features without missing values, and up to 10 classes, we show that our method clearly outperforms boosted trees and performs on par with complex state-of-the-art AutoML systems with up to 230× speedup. This increases to a 5 700× speedup when using a GPU. We also validate these results on an additional 67 small numerical datasets from OpenML. We provide all our code, the trained TabPFN, an interactive browser demo and a Colab notebook at https://github.com/automl/TabPFN.","description_state":"present","introduced_year":null,"introduced_by":{"title":"TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second","paper":"/paper/meta-learning-a-real-time-tabular-automl","first_author":"Noah Hollmann","n_authors":4,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/meta-learning-a-real-time-tabular-automl"},"source":{"url":"https://arxiv.org/abs/2207.01848v6","title":"TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Deep Tabular Learning","url":"/methods/category/deep-tabular-learning","pwc_aliases":[]}],"n_papers_tagged":26,"archive_num_papers":26,"papers_newest_first":[{"paper":"/paper/contexttab-a-semantics-aware-tabular-in","title":"ConTextTab: A Semantics-Aware Tabular In-Context Learner","date":"2025-06-12","arxiv_id":"2506.10707","n_code_links":1,"syntology":{"ran":2,"of":6,"unverified":4,"pointer_only":0}},{"paper":null,"title":"On the Robustness of Tabular Foundation Models: Test-Time Attacks and In-Context Defenses","date":"2025-06-03","arxiv_id":"2506.02978","n_code_links":0,"syntology":null},{"paper":"/paper/tabpfn-one-model-to-rule-them-all","title":"TabPFN: One Model to Rule Them All?","date":"2025-05-26","arxiv_id":"2505.20003","n_code_links":1,"syntology":null},{"paper":null,"title":"Realistic Evaluation of TabPFN v2 in Open Environments","date":"2025-05-22","arxiv_id":"2505.16226","n_code_links":0,"syntology":null},{"paper":null,"title":"Tabular foundation model to detect empathy from visual cues","date":"2025-04-15","arxiv_id":"2504.10808","n_code_links":0,"syntology":null},{"paper":null,"title":"A Closer Look at TabPFN v2: Strength, Limitation, and Extension","date":"2025-02-24","arxiv_id":"2502.17361","n_code_links":0,"syntology":null},{"paper":null,"title":"TabPFN Unleashed: A Scalable and Effective Solution to Tabular Classification Problems","date":"2025-02-04","arxiv_id":"2502.02527","n_code_links":0,"syntology":null},{"paper":"/paper/transformers-boost-the-performance-of","title":"Transformers Boost the Performance of Decision Trees on Tabular Data across Sample Sizes","date":"2025-02-04","arxiv_id":"2502.02672","n_code_links":1,"syntology":null},{"paper":"/paper/the-tabular-foundation-model-tabpfn","title":"The Tabular Foundation Model TabPFN Outperforms Specialized Time Series Forecasting Models Based on Simple Features","date":"2025-01-06","arxiv_id":"2501.02945","n_code_links":1,"syntology":{"ran":3,"of":5,"unverified":2,"pointer_only":0}},{"paper":null,"title":"Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data","date":"2024-11-15","arxiv_id":"2411.10634","n_code_links":0,"syntology":null},{"paper":null,"title":"A Survey on Deep Tabular Learning","date":"2024-10-15","arxiv_id":"2410.12034","n_code_links":0,"syntology":null},{"paper":null,"title":"AnnotatedTables: A Large Tabular Dataset with Language Model Annotations","date":"2024-06-24","arxiv_id":"2406.16349","n_code_links":0,"syntology":null},{"paper":"/paper/large-scale-transfer-learning-for-tabular","title":"Large Scale Transfer Learning for Tabular Data via Language Modeling","date":"2024-06-17","arxiv_id":"2406.12031","n_code_links":2,"syntology":{"ran":13,"of":15,"unverified":2,"pointer_only":0}},{"paper":null,"title":"Grapevine Disease Prediction Using Climate Variables from Multi-Sensor Remote Sensing Imagery via a Transformer Model","date":"2024-06-11","arxiv_id":"2406.07094","n_code_links":0,"syntology":null},{"paper":null,"title":"Tokenize features, enhancing tables: the FT-TABPFN model for tabular classification","date":"2024-06-11","arxiv_id":"2406.06891","n_code_links":0,"syntology":null},{"paper":null,"title":"Retrieval & Fine-Tuning for In-Context Tabular Models","date":"2024-06-07","arxiv_id":"2406.05207","n_code_links":0,"syntology":null},{"paper":"/paper/tabpfgen-tabular-data-generation-with-tabpfn","title":"TabPFGen -- Tabular Data Generation with TabPFN","date":"2024-06-07","arxiv_id":"2406.05216","n_code_links":1,"syntology":null},{"paper":"/paper/why-in-context-learning-transformers-are","title":"Fine-tuned In-Context Learning Transformers are Excellent Tabular Data Classifiers","date":"2024-05-22","arxiv_id":"2405.13396","n_code_links":1,"syntology":{"ran":4,"of":4,"unverified":0,"pointer_only":0}},{"paper":"/paper/interpretable-machine-learning-for-tabpfn","title":"Interpretable Machine Learning for TabPFN","date":"2024-03-16","arxiv_id":"2403.10923","n_code_links":1,"syntology":{"ran":8,"of":12,"unverified":4,"pointer_only":0}},{"paper":"/paper/tunetables-context-optimization-for-scalable","title":"TuneTables: Context Optimization for Scalable Prior-Data Fitted Networks","date":"2024-02-17","arxiv_id":"2402.11137","n_code_links":2,"syntology":{"ran":10,"of":17,"unverified":7,"pointer_only":0}},{"paper":null,"title":"In-Context Data Distillation with TabPFN","date":"2024-02-10","arxiv_id":"2402.06971","n_code_links":0,"syntology":null},{"paper":"/paper/is-mamba-capable-of-in-context-learning","title":"Is Mamba Capable of In-Context Learning?","date":"2024-02-05","arxiv_id":"2402.03170","n_code_links":1,"syntology":{"ran":5,"of":8,"unverified":3,"pointer_only":8}},{"paper":"/paper/mothernet-a-foundational-hypernetwork-for","title":"MotherNet: Fast Training and Inference via Hyper-Network Transformers","date":"2023-12-14","arxiv_id":"2312.08598","n_code_links":1,"syntology":{"ran":1,"of":1,"unverified":0,"pointer_only":0}},{"paper":null,"title":"Scaling TabPFN: Sketching and Feature Selection for Tabular Prior-Data Fitted Networks","date":"2023-11-17","arxiv_id":"2311.10609","n_code_links":0,"syntology":null},{"paper":null,"title":"Fine-Tuning the Retrieval Mechanism for Tabular Deep Learning","date":"2023-11-13","arxiv_id":"2311.07343","n_code_links":0,"syntology":null},{"paper":"/paper/meta-learning-a-real-time-tabular-automl","title":"TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second","date":"2022-07-05","arxiv_id":"2207.01848","n_code_links":7,"syntology":{"ran":1,"of":4,"unverified":3,"pointer_only":1}}],"papers_shown":26,"tasks":[{"task":"/task/in-context-learning","name":"In-Context Learning","papers":16},{"task":"/task/tabular-classification","name":"tabular-classification","papers":6},{"task":"/task/automl","name":"AutoML","papers":3},{"task":"/task/bayesian-inference","name":"Bayesian Inference","papers":3},{"task":"/task/meta-learning","name":"Meta-Learning","papers":3},{"task":"/task/transfer-learning","name":"Transfer Learning","papers":3},{"task":"/task/classification-1","name":"Classification","papers":2},{"task":"/task/language-modeling","name":"Language Modeling","papers":2},{"task":"/task/language-modelling","name":"Language Modelling","papers":2},{"task":"/task/prompt-engineering","name":"Prompt Engineering","papers":2},{"task":"/task/retrieval","name":"Retrieval","papers":2},{"task":"/task/feature-selection","name":"feature selection","papers":2},{"task":"/task/all","name":"All","papers":1},{"task":"/task/computational-efficiency","name":"Computational Efficiency","papers":1},{"task":"/task/data-augmentation","name":"Data Augmentation","papers":1},{"task":"/task/data-valuation","name":"Data Valuation","papers":1},{"task":"/task/deep-learning","name":"Deep Learning","papers":1},{"task":"/task/denoising","name":"Denoising","papers":1},{"task":"/task/density-estimation","name":"Density Estimation","papers":1},{"task":"/task/disease-prediction","name":"Disease Prediction","papers":1}],"tasks_shown":20,"n_tasks":44,"usage_by_year":[{"year":"2022","papers":1},{"year":"2023","papers":3},{"year":"2024","papers":13},{"year":"2025","papers":9}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/tabpfn"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}