#216216th of 242 in ML frameworks and librariesFor data and documents, data analysis
orbit
Python library for Bayesian time series forecasting with a fit-predict API
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 2.1k
- License
- custom license
- Last commit
- May 2026
- Last release
- May 2026
- Language
- Python
2,041 on 25 Apr 20262,083 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Orbit is a Python package from Uber for probabilistic time series forecasting. It exposes an initialize-fit-predict interface over Stan-based models: ETS, LGT, DLT and KTR. Estimation can use MCMC, MAP or variational inference. It is a statistical forecasting library, not an LLM tool.
Who it is for: Data scientists wanting probabilistic forecasts in Python
Strengths
- Four model families (ETS, LGT, DLT, KTR) behind one fit/predict interface
- Three estimation modes: MCMC, MAP and variational inference
- Installable from PyPI or conda-forge, with Colab tutorials and M3 backtest example
Weaknesses
- Requires cmdstanpy for Bayesian sampling, adding a Stan toolchain dependency
- README says some experimental code has APIs subject to change
- Covers time series forecasting only, with no deep learning models
- Default page is the dev branch; stable code lives on master
What it needs
- Docker
- Needs cmdstanpy, Stan, Pyro
- Models: ETS, LGT, DLT, KTR
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in ML frameworks and libraries
See all 242| Rank | Project | Score |
|---|---|---|
| 1 | transformersModel-definition library for pretrained text, vision, audio and multimodal models | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 94 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |