nixtla
TimeGPT-1: production ready pre-trained Time Series Foundation Model for forecasting and anomaly detection. Generative pretrained transformer for time series trained on over 100B data points. It's capable of accurately predicting various domains such as retail, electricity, finance, and IoT with just a few lines of code 🚀.
- stars
- 4,008
- 30-day movement
- +41/day
- Related entries
- 60
- Connections
- 2
Nixtla hosts TimeGPT-1, a pre-trained generative transformer foundation model for time series forecasting and anomaly detection, trained on over 100B data points. The repository provides access to the model for producing forecasts across domains like retail, electricity, finance, and IoT with a few lines of code.
You want production-ready time series forecasting or anomaly detection from a pre-trained foundation model without training your own.
Use it to
- Forecast time series across retail, electricity, finance, and IoT domains
- Detect anomalies in time series data
- Generate forecasts with minimal code via a pre-trained model
- Experiment with TimeGPT in a reproducible devcontainer environment
- Integrate forecasting into agentic AI workflows
For Data scientists and ML engineers needing time series forecasting
- Role
- agent-app
- Language
- Jupyter Notebook
- Licence
- custom (licence file present)
- Forks
- 336
- Open issues
- 44
- Last push
- 2026-09-15
- Latest release
- v0.1.0 · 2023-08-05