Nixtla/mlforecast

Scalable machine 🤖 learning for time series forecasting.

76
/ 100
Verified

This tool helps businesses and analysts predict future trends using historical time series data. You input multiple series of past observations, like sales figures or stock prices, and it outputs predictions for future values. It's designed for data scientists, operations managers, and anyone needing accurate, scalable forecasts for business planning or resource allocation.

1,186 stars. Used by 2 other packages. Actively maintained with 21 commits in the last 30 days. Available on PyPI.

Use this if you need to generate fast, accurate machine learning-powered forecasts for many time series simultaneously, even with very large datasets.

Not ideal if you prefer traditional statistical forecasting methods over machine learning, or only need to forecast a single time series without needing to scale.

demand-forecasting financial-forecasting inventory-planning operations-analytics data-science
Maintenance 20 / 25
Adoption 12 / 25
Maturity 25 / 25
Community 19 / 25

How are scores calculated?

Stars

1,186

Forks

122

Language

Python

License

Apache-2.0

Last pushed

Mar 12, 2026

Commits (30d)

21

Dependencies

8

Reverse dependents

2

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Nixtla/mlforecast"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.