Skip to main content
Synthefy Tabular predicts continuous values from tabular data. You give it some labeled rows as examples, and it predicts on new rows — no training, no fine-tuning required. GitHub  ·  🤗 Hugging Face There are two ways to use it:

Run locally

Install the Python package and run inference on your own machine or server.

Call the API

Send a request to our hosted endpoint. No setup, no GPU required.

​
Local

Install the package: 
pip install synthefy-tabular
Model weights download automatically from Hugging Face on first use — no API key needed.

​
Your first prediction

from synthefy_tabular import SynthefyTabularRegressor

X_train = [[0.0, 1.0], [1.0, 0.0], [0.5, 0.5], [0.2, 0.8]]
y_train = [0.1, 0.9, 0.5, 0.3]
X_test  = [[0.3, 0.7], [0.8, 0.2]]

model = SynthefyTabularRegressor()
model.fit(X_train, y_train)
pred = model.predict(X_test)
print(pred)

[0.36 0.74]
Or use the one-shot helper: 
from synthefy_tabular import predict

pred = predict(X_train, y_train, X_test, task="regression")

[0.36 0.74]

​
With a DataFrame

import pandas as pd
from synthefy_tabular import SynthefyTabularRegressor

df = pd.read_csv("data.csv")
target_col = "price"
feature_cols = [c for c in df.columns if c != target_col]

train = df.sample(frac=0.8, random_state=42)
test  = df.drop(train.index)

model = SynthefyTabularRegressor()
model.fit(train[feature_cols].values, train[target_col].values)
predictions = model.predict(test[feature_cols].values)

​
Handle missing values

You don’t need to fill in missing values beforehand — the model handles them automatically. 
import numpy as np
from synthefy_tabular import SynthefyTabularRegressor

X_train = [[0.0, 1.0], [1.0, np.nan], [0.5, 0.5], [np.nan, 0.8]]
y_train = [0.1, 0.9, 0.5, 0.3]
X_test  = [[0.3, 0.7], [0.8, np.nan]]

model = SynthefyTabularRegressor()
model.fit(X_train, y_train)
print(model.predict(X_test))

[0.37 0.69]
The model uses a GPU automatically if one is available, and falls back to CPU otherwise.

​
API

The hosted API runs the same model on our infrastructure — no installation or GPU required. Set up your API key in the API key guide.

​
Make a request

Send your labeled rows (X_train, y_train) and the rows you want to predict (X_test) in a single call: 
curl -X POST https://model-3m5j7y9w.api.baseten.co/environments/production/predict \
  -H "Authorization: Api-Key $SYNTHEFY_TABULAR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "X_train": [[0.0, 1.0], [1.0, 0.0], [0.5, 0.5], [0.2, 0.8]],
    "y_train": [0.1, 0.9, 0.5, 0.3],
    "X_test":  [[0.3, 0.7], [0.8, 0.2]]
  }'

​
Response

{
  "task": "regression",
  "predictions": [0.36, 0.74],
  "usage": {
    "input_tokens": 16,
    "output_tokens": 2,
    "total_tokens": 18
  }
}
predictions contains one value per row in X_test. The usage field follows the OpenAI token counting convention — input_tokens counts every non-null value sent, output_tokens is one per predicted row.
The first request after the model has been idle may take ~90 seconds while it warms up. Subsequent requests return in 1–2 seconds.

​
Resources

White paper

Coming soon.

Product page

Coming soon.

GitHub

Source code for training, inference, and evaluation.

Hugging Face

Pretrained model weights.