Synthefy-Nori-V1 — Replaces XGBoost

→
SynthefySynthefy

New Release · Synthefy-Nori-V1

Train nothing.Predict anything.

The foundation model for tables. Fully open source.

Get started 🤗 Get the weightsView on GitHubSee benchmarksTry through Baseten
  • #1 mean R² across 96 regression datasets
  • Beats tuned XGBoost on 9 of 13 TabArena tasks
  • #1 in CRPS & beta-energy on ScoringBench
In-context learning: labeled context rows go into Synthefy-Nori-V1, predictions come out in one forward pass — no training, no tuning

Meet Synthefy-Nori

Delete the loop. Keep the predictions.

Your labeled rows are the context, and the predictions come back in a single forward pass. The model handles preprocessing, high dimensionality, and skewed targets on its own.

Without Synthefy: pick a model, train and tune, validate, then retrain as data drifts. With Synthefy: one forward pass from your data to predictions.
  1. 01

    Hand it your labeled rows

    Pass your training table — X_train and y_train — straight into the call as context. No gradient updates, no training loop, no knobs to turn.

  2. 02

    One forward pass

    A single predict() runs your rows through the model once. Missing values, redundant columns, and skewed targets are handled internally.

  3. 03

    Predictions out

    No validation sweep, no model-versioning sprawl. When the data drifts, you send the new rows as context — there is nothing to retrain.

This is the entire API.

PYTHON
from synthefy_nori import NoriRegressor
 
model = NoriRegressor()
model.fit(X_train, y_train)
predictions = model.predict(X_test)
BasetenNori is available as an API through Baseten.

Benchmark proof

The best tabular model in the world.

Synthefy-Nori-V1 was evaluated on 96 regression datasets from three independent sources — TabArena, TALENT, and OpenML-Reg. Same train/test splits, same preprocessing, same hardware for every model. Higher R² is better.

0training runsyour labeled rows are the context
#1mean R²across 96 regression datasets
9/13beats completely tunedgradient boosting models
6Mparametersa tenth of TabPFN-3
Highest aggregate R² of any tabular foundation model: Synthefy-Nori-V1 at 0.7507, ahead of TabPFN-3, TabPFN-2.6, TabPFN-2.5, TabICLv2, and LimiX-2M

Highest aggregate R² of any tabular foundation model.

R² measures how much of the variation in the target a model explains — higher is better, 1.0 is perfect. Averaged across all 96 datasets, Synthefy-Nori-V1 leads TabPFN-3, the strongest prior tabular foundation model, at a tenth of the size.

Per-dataset accuracy on TabArena: Synthefy-Nori-V1 wins 9 of 13 regression datasets against tuned XGBoost and LightGBM (AutoGluon best-quality, 4h), and on the four it loses the gap is under 2%

Wins where gradient boosting is strongest.

TabArena skews toward larger, modern datasets — gradient boosting’s home turf. With zero tuning, Synthefy-Nori wins 9 of 13 regression datasets against tuned XGBoost and LightGBM (AutoGluon’s best-quality preset, a 4-hour budget) — and on the four it loses, the margin is under 2%.

A tenth of the size of the model it beats: Synthefy-Nori-V1 at 6M parameters versus TabPFN-3 at 58.3M, with higher mean R²

A tenth of the size of the model it beats.

6M parameters versus TabPFN-3 at 58.3M — with higher mean R². The diamonds regression (16K rows) runs end to end in ~2.8 seconds on a single GPU.

Accuracy isn't the only axis. On ScoringBench, the independent leaderboard for probabilistic forecasting, Synthefy-Nori-V1 ranks #1 in CRPS and across the beta-energy family of proper scoring rules — its predictions aren't just accurate on average, they're well-calibrated across the full distribution.

Teaser — Thinking Mode

It thinks before it predicts.

Thinking Mode decides how to process each dataset before predicting — augmentations, normalizations, preprocessing — with no human in the loop. The gains land on the large, hard datasets and compound in aggregate, lifting mean R² to 0.7531.

Thinking extends the lead: Synthefy-Nori-V1 + Thinking reaches 0.7531 mean R², above the base model and well ahead of TabPFN-3

The payoff

No more madness.

The months you spend on the pipeline — EDA, data engineering, feature selection, training, tuning — collapse into two function calls. Here's what leaves your workflow for good:

Exploratory data analysisFeature engineeringModel selectionHyperparameter searchCross-validationDrift-driven retrainingMLOps glue code

The tuning ritual evaporates

No training loops, no learning-rate sweeps, no early-stopping callbacks. There are no knobs to turn — the model configures its own preprocessing per dataset.

Stop cleaning data to keep the model happy

Missing values, noisy labels, redundant columns, heavy tails — Synthefy-Nori was pretrained on synthetic data deliberately built to contain all of it. Hand it the raw rows.

Drift becomes a non-event

Drift used to mean spinning up a training run. With in-context learning it means sending the new rows in as context. No retraining, no model-versioning sprawl.

A closer look

When Nori wins, it wins big.

Across 96 datasets the two models usually tie, which keeps the average margin small. But where either model has a decisive edge, Nori lands the most wins — and the largest — on real, public datasets, at a tenth of the size. And on the small-to-mid tables it’s built for, it returns predictions faster too.

The datasets where Synthefy-Nori-V1 beats TabPFN-3 by more than noise: Job Profitability 0.14→0.41, socmob 0.78→0.89, SAT11-HAND 0.70→0.78, WLAN RSSI 0.89→0.94, sulfur 0.88→0.91 and more — 8 of the 11 decisive matchups, by the widest margins

When it wins, it wins big — and here’s where.

Most datasets are a tie, which keeps the average margin small. But of the 11 datasets where either model has a decisive edge (>0.02 R²), Nori takes 8 — by the widest margins. The standout is Job Profitability, where it lifts R² from 0.14 to 0.41, tripling the explained variance. socmob and sulfur also win under a second independent benchmark suite, so these aren’t harness flukes — every dataset is public.

Median wall-clock latency on the 48 benchmark datasets up to 100k cells: Synthefy-Nori-V1 (6M params) is faster than TabPFN-3 (58.3M) in every size band

Faster on the tables it’s built for.

On those small-to-mid tables, Nori returns predictions in roughly a second — faster than TabPFN-3 in every size band, at 6M parameters versus 58.3M. No training run, no cluster: one library call on a single GPU. Past ~100k cells, a quick gradient-boosted model still wins — we’d rather be straight about that.

Security & compliance

Enterprise-grade by default.

The managed Nori API is delivered on Baseten's infrastructure, which is SOC 2 Type II certified and retains none of your inputs or outputs by default. Prefer to self-host? Nori is open source under Apache 2.0 and free to run in your own environment.

  • SOC 2 Type II
  • HIPAA
  • GDPR
  • Zero Data Retention

Quickstart

From table to prediction in three lines.

Install the package, point it at your data, and you’re predicting. The first call pulls the weights from Hugging Face and caches them — there’s nothing to train, tune, or configure.

Install

pip install synthefy-nori

Run

PYTHON
from sklearn.datasets import load_diabetes
from sklearn.model_selection import train_test_split
from synthefy_nori import NoriRegressor
 
X, y = load_diabetes(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
 
model = NoriRegressor()              # downloads weights on first use
model.fit(X_train, y_train)          # stores your labeled rows as context
predictions = model.predict(X_test)  # one forward pass — no training

Fully open source — Apache 2.0

Every table you own is a prediction you haven't made yet.

Code on GitHub, weights on Hugging Face — free and open source. Point it at the data you already have and see what it predicts.

🤗 Get the weightsGitHub Read the docsJoin DiscordTalk to us