---
id: gluonts
version: "0.17.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# gluonts — Probabilistic time series modeling in Python.

License: permissive · Maintenance: active · Downloads: 8.5M/mo

## What it is and what it does

GluonTS is a Python package for building and training probabilistic time series forecasting models using deep learning. It provides high-level estimators that learn patterns in historical time series data and generate full probability distributions over future values, rather than single-point predictions. The package handles data loading, train-test splitting, model training, and forecast generation through a unified API.

Typical usage involves loading time series data into a dataset object, splitting it for training and evaluation, instantiating an estimator with forecast parameters, training on historical data, and then generating probabilistic predictions with confidence intervals. The package depends on numpy, pandas, pydantic, toolz, tqdm, and typing-extensions, making it compatible with standard data science workflows.

Use it for:

- Train a model on historical sales data to forecast demand with prediction intervals for inventory planning
- Build probabilistic forecasts for energy consumption to quantify uncertainty in grid load predictions
- Generate time series predictions for financial metrics where understanding forecast confidence is critical
- Benchmark multiple deep learning architectures on your own time series dataset using available estimators
- Integrate pretrained models for zero-shot forecasting on new time series without retraining

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

GluonTS provides deep learning models for probabilistic time series forecasting, built on PyTorch, enabling you to train and deploy models that generate probability distributions over future values rather than point estimates.

Yes. GluonTS is actively maintained, has no known vulnerabilities, installs with low friction, and is licensed permissively. It is the right choice if you need probabilistic deep learning forecasts with confidence intervals and want a mature, well-documented framework. Install it if your use case requires neural time series models; skip it if you need only classical statistical forecasting or simple point predictions.

## Install

pip install gluonts
uv add gluonts
poetry add gluonts

## Installing gluonts

Before you install: Low friction install with a pure-Python wheel. Active maintenance with a recent release (14 days old) and strong repository signals (5224 stars, last commit 2026-07-31). Requires Python 3.10–3.14; the package explicitly supports current Python versions.

License in practice: Apache-2.0 permissive license allows commercial and private use with minimal restrictions, making it suitable for most production and research contexts.

Quickstart:

pip install gluonts

import pandas as pd
from gluonts.dataset.pandas import PandasDataset
from gluonts.torch import DeepAREstimator

df = pd.read_csv('data.csv', index_col=0, parse_dates=True)
dataset = PandasDataset(df, target='value')
model = DeepAREstimator(prediction_length=12, freq='M').train(dataset)
forecasts = list(model.predict(dataset))

Requires Python 3.10–3.14. PyTorch must be available as a runtime dependency for model training.

Verify before relying:

- Whether core functionality requires the [torch] extra or if CPU-only forecasting is viable without it
- Performance characteristics and scalability limits for large time series datasets
- Availability and quality of pretrained Chronos models mentioned in the description

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 8.5M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

time series forecasting deep learning, probabilistic forecasting models, neural time series prediction, time series deep learning pytorch, probabilistic time series modeling, time-series-forecasting, deep-learning, probabilistic-modeling

[View on SkillFed](https://skillfed.io/packages/gluonts) · [View on PyPI](https://pypi.org/project/gluonts/)