Skip to main content

​
How to Use TabPFN

TabPFN can be used in two ways - depending on your workflow and compute setup. Whether you’re experimenting locally, integrating into a pipeline, or deploying at scale, there’s a path that fits.

​
1. Open Source Package

Best for: researchers, ML practitioners, and engineers with local GPU or cluster access. TabPFN is available as an open-source Python package on GitHub. It provides the full model locally, giving you complete control over inference and experimentation. 
pip install tabpfn

from tabpfn import TabPFNClassifier

model = TabPFNClassifier()
model.fit(X_train, y_train)
preds = model.predict(X_test)
GPU recommended (e.g., NVIDIA T4 at minimum) or A100/H100 for optimal inference speed.

​
2. TabPFN API

Best for: data teams and developers who want TabPFN’s performance without managing infrastructure. The Prior Labs API provides cloud-hosted access to TabPFN-2.5 - we handle all GPU compute, scaling, and model versioning. You can use it either through a REST API or our Python SDK.
You must sign up and generate an API key before making any requests. Requests without valid authentication headers will be rejected.

​
Using the REST API

First, define your dataset paths and authentication details. This allows you to reuse them across fit and predict calls. 
import os, json, requests

# Define your file paths
train_path = "train.csv"   # path to your training dataset
test_path = "test.csv"     # path to your test dataset

# Get your API key from the environment
api_key = os.getenv("PRIORLABS_API_KEY")

headers = {"Authorization": f"Bearer {api_key}"}
Next, upload your training dataset to the /v1/fit endpoint. The API automatically preprocesses the data. 
payload = {"task": "classification"}

files = {
    "data": (None, json.dumps(payload), "application/json"),
    "dataset_file": (train_path, open(train_path, "rb")),
}

fit_response = requests.post(
    "https://api.priorlabs.ai/v1/fit",
    headers=headers,
    files=files,
)

model_id = fit_response.json().get("model_id")
print(f"✅ Model trained: {model_id}")
Once fitted your model using in-context learning, send your test dataset to the /v1/predict endpoint using the returned model_id. You’ll receive predictions and/or probabilities in JSON format. 
payload = {
    "task": "classification",
    "model_id": model_id,
}

files = {
    "data": (None, json.dumps(payload), "application/json"),
    "file": (test_path, open(test_path, "rb")),
}

predict_response = requests.post(
    "https://api.priorlabs.ai/v1/predict",
    headers=headers,
    files=files,
)

print("✅ Predictions:")
print(json.dumps(predict_response.json(), indent=2))

API Reference

You can follow the step-by-step onboarding instructions in the API Reference, which also includes detailed endpoint descriptions and authentication examples.

​
Using the Python SDK

import tabpfn_client
from tabpfn_client import TabPFNClassifier, TabPFNRegressor

# Authenticate using your Prior Labs token
tabpfn_client.set_access_token("YOUR_API_KEY")

# Use TabPFN like any scikit-learn model
model = TabPFNClassifier()
model.fit(X_train, y_train)

# Get predictions
preds = model.predict(X_test)

API Reference

See the full OpenAPI spec, endpoint details, and request/response formats.

SDKs

View available SDKs for Python and JavaScript, usage instructions, and code samples.
⌘I