TempoPFN: Synthetic Pre-Training of Linear RNNs for Zero-Shot Time Series Forecasting

preprint GIFT-Eval github License: Apache 2.0


TempoPFN introduced in TempoPFN: Synthetic Pre-Training of Linear RNNs for Zero-Shot Time Series Forecasting, is a univariate time series foundation model pretrained entirely on synthetic data. It delivers top-tier zero-shot forecasting accuracy while remaining fully reproducible and free from real-data leakage.

Built on a Linear RNN (GatedDeltaProduct) backbone, TempoPFN performs end-to-end forecasting without patching or windowing. Its design enables fully parallelizable training and inference while maintaining stable temporal state-tracking across long sequences. The GatedDeltaProduct architecture is based on DeltaProduct, extended with state-weaving for time series forecasting. For detailed information about the architecture and custom modifications, see src/models/gated_deltaproduct/README.md.

This repository includes the pretrained 38M parameter model (models/checkpoint_38M.pth), all training and inference code, and the complete synthetic data generation pipeline used for pretraining.

✨ Why TempoPFN?

  • High Performance, No Real Data: Achieves top-tier competitive results on GIFT-Eval, outperforming all existing synthetic-only approaches and surpassing the vast majority of models trained on real-world data. This ensures full reproducibility and eliminates benchmark leakage.
  • Parallel and Efficient: The linear recurrence design enables full-sequence parallelization. This gives us the best of both worlds: the linear efficiency of an RNN, but with the training parallelism of a Transformer.
  • Open and Reproducible: Includes the full synthetic data pipeline, configurations, and scripts to reproduce training from scratch.
  • State-Tracking Stability: The GatedDeltaProduct recurrence and state-weaving mechanism preserve temporal continuity and information flow across long horizons, improving robustness without non-linear recurrence.

TempoPFN Overview

βš™οΈ Installation

Note on Model Weights: This repository uses Git LFS to store the model checkpoint (.pth file). You must have Git LFS installed to clone the repository correctly.

# Install Git LFS (e.g., on Ubuntu)
sudo apt-get install git-lfs
git lfs install
  1. Clone the repository:
    git clone https://huggingface.co/AutoML-org/TempoPFN
    cd TempoPFN
  1. Set up the environment:
python -m venv venv && source venv/bin/activate

# 1. Install PyTorch version matching your CUDA version
# Example for CUDA 12.8:
pip install torch --index-url https://download.pytorch.org/whl/cu128

# 2. Clone the Hugging Face repository
git clone https://huggingface.co/AutoML-org/TempoPFN
cd TempoPFN

# 3. Set up the environment
python3.12 -m venv venv & source venv/bin/activate
export PYTHONPATH=$PWD

# 4. Install PyTorch version matching your CUDA version
pip install torch --index-url https://download.pytorch.org/whl/cu128

# 5. Install dependencies
pip install .
pip install .[dev]

# 4. Run the Quick Start Script 
python examples/quick_start_tempo_pfn.py

# 5. Alternatively, you can run the Notebook version
jupyter notebook examples/quick_start_tempo_pfn.ipynb

Hardware & Performance Tips

GPU Required: Inference requires a CUDA-capable GPU with a matching PyTorch version installed. Tested on NVIDIA A100/H100.

First Run: The first inference for a new sequence length will be slow due to Triton kernel compilation. Subsequent runs will be fast.

Cache Tip: If using a network filesystem, prevent slowdowns by routing caches to a local directory (like /tmp) before running:

LOCAL_CACHE_BASE="${TMPDIR:-/tmp}/tsf-$(date +%s)"
mkdir -p "${LOCAL_CACHE_BASE}/triton" "${LOCAL_CACHE_BASE}/torchinductor"
export TRITON_CACHE_DIR="${LOCAL_CACHE_BASE}/triton"
export TORCHINDUCTOR_CACHE_DIR="${LOCAL_CACHE_BASE}/torchinductor"

python examples/quick_start_tempo_pfn.py

πŸš‚ Training

All training and model parameters are controlled via YAML files in configs/.

# Single-GPU (Debug)
torchrun --standalone --nproc_per_node=1 src/training/trainer_dist.py --config ./configs/train.yaml

# Multi-GPU (e.g., 8 GPUs)
torchrun --standalone --nproc_per_node=8 src/training/trainer_dist.py --config ./configs/train.yaml

πŸ’Ύ Synthetic Data Generation

A core contribution of this work is our open-source synthetic data pipeline, located in src/synthetic_generation/. It combines diverse generators with a powerful augmentation cascade.

Generators Used:

  • Adapted Priors: ForecastPFN, KernelSynth, GaussianProcess (GP), and CauKer (Structural Causal Models).
  • Novel Priors: SDE (a flexible regime-switching Ornstein-Uhlenbeck process), Sawtooth, StepFunction, Anomaly, Spikes, SineWave, and Audio-Inspired generators (Stochastic Rhythms, Financial Volatility, Network Topology, Multi-Scale Fractals).

You can easily generate your own data by installing the development dependencies and instantiating a generator wrapper. See examples/generate_synthetic_data.py for a minimal script, or inspect the generator code in src/synthetic_generation/.

🀝 License

This project is licensed under the Apache 2.0 License. See the LICENSE file for details. This permissive license allows for both academic and commercial use.

πŸ“š Citation

If you find TempoPFN useful in your research, please consider citing our paper:

@misc{moroshan2025tempopfn,
  title={TempoPFN: Synthetic Pre-training of Linear RNNs for Zero-Shot Time Series Forecasting}, 
  author={Vladyslav Moroshan and Julien Siems and Arber Zela and Timur Carstensen and Frank Hutter},
  year={2025},
  eprint={2510.25502},
  archivePrefix={arXiv},
  primaryClass={cs.LG}
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Space using AutoML-org/TempoPFN 1