#9797th of 206 in ML frameworks and librariesFor data and documents, data analysis
nixtla
Python client SDK for the hosted TimeGPT time series forecasting API
- Stars
- 4k
- License
- custom license
- Last commit
- Nov 2026
- Last release
- Sep 2026
- Language
- Jupyter Notebook
3,832 on 22 Apr 20264,047 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Nixtla is the Python SDK (NixtlaClient) for TimeGPT, a closed-source transformer foundation model for time series. It sends data to the hosted model for zero-shot forecasting, anomaly detection, fine-tuning, cross-validation and prediction intervals, with exogenous variables and multi-series support. The model does not run locally, so an API key is required.
Who it is for: Data scientists and engineers forecasting time series via a hosted API
Strengths
- Forecast or detect anomalies in a few lines using pandas DataFrames
- Zero-shot inference, plus fine-tuning, cross-validation and prediction intervals
- Handles exogenous variables, multiple series and irregular timestamps
- Docs cover Spark, Dask and Ray scaling; Snowflake install script included
Weaknesses
- Model is closed source; the repo holds only the client SDK
- Every call needs an API key; access starts with a free trial
- detect_anomalies() is removed in 1.0 with no drop-in replacement
What it needs
- no GPU
- Needs Nixtla TimeGPT API key, pandas
- Models: TimeGPT
What changed
- v0.9.0 ↗ (opens in a new tab)Breaking
“Breaking: detectanomaliesonline() gained finetunedmodelid and modelparameters mid-signature, to match the parameter order of forecast() and crossvalidation(). Callers passing arguments positionally p…”
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 206| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |