# TimeGPT Foundational model for time series forecasting and anomaly detection

## Docs

- [Key Concepts](https://nixtla.io/docs/about/key-concepts.md): Understanding the foundations of time series forecasting with TimeGPT
- [Privacy Notice](https://nixtla.io/docs/about/privacy-notice.md): Details on how Nixtla collects, uses, and protects your personal information.
- [Nixtla](https://nixtla.io/docs/about/sub-categoria.md): About us
- [Terms and Conditions](https://nixtla.io/docs/about/terms-and-conditions.md): Terms and conditions for using Nixtla Services.
- [Add Exogenous Variables](https://nixtla.io/docs/anomaly_detection/exogenous_variables.md): Learn how to improve anomaly detection by incorporating external factors.
- [Quickstart](https://nixtla.io/docs/anomaly_detection/historical_anomaly_detection.md): Get started with TimeGPT's historical anomaly detection capabilities.
- [Controlling the Anomaly Detection Process](https://nixtla.io/docs/anomaly_detection/real-time/adjusting_detection.md): Learn how to tune TimeGPT's anomaly detection parameters for optimal accuracy. Step-by-step guide to adjusting detection_size, level, confidence intervals, and fine-tuning strategies with Python code examples.
- [Online (Real-Time) Anomaly Detection](https://nixtla.io/docs/anomaly_detection/real-time/introduction.md): Learn how to detect anomalies in real-time streaming data using TimeGPT's detect_anomalies_online method. Complete Python tutorial with code examples for monitoring server logs, IoT sensors, and live data streams.
- [Local vs Global Anomaly Detection](https://nixtla.io/docs/anomaly_detection/real-time/univariate_multivariate.md): Compare local vs global anomaly detection methods for time series. Learn when to use univariate detection for independent metrics vs multivariate detection for correlated server data with Python examples.
- [Delete Fine-tuned Model](https://nixtla.io/docs/api-reference/delete-fine-tuned-model.md): Delete a previously saved finetuned model. It takes the ID of the model that you want to delete as a path parameter.
- [Foundational Time Series Model Multi Series](https://nixtla.io/docs/api-reference/foundational-time-series-model-multi-series.md): Based on the provided data, this endpoint predicts the future values of multiple time series at once. It takes a JSON as an input containing information like the series frequency and historical data. (See below for a full description of the parameters.) The response contains the predicted values for…
- [Foundational Time Series Model Multi Series Anomaly Detector](https://nixtla.io/docs/api-reference/foundational-time-series-model-multi-series-anomaly-detector.md): Based on the provided data, this endpoint detects the anomalies in the historical perdiod of multiple time series at once. It takes a JSON as an input containing information like the series frequency and historical data. (See below for a full description of the parameters.) The response contains a f…
- [Foundational Time Series Model Multi Series Cross Validation](https://nixtla.io/docs/api-reference/foundational-time-series-model-multi-series-cross-validation.md): Perform Cross Validation for multiple series
- [Foundational Time Series Model Multi Series Finetuning](https://nixtla.io/docs/api-reference/foundational-time-series-model-multi-series-finetuning.md): Fine-tune the large time model to your data and save it for later use. It takes a JSON as an input containing information like the series frequency and historical data. (See below for a full description of the parameters.) The response contains the ID of the finetuned model, which you can provide in…
- [Foundational Time Series Model Multi Series Historic](https://nixtla.io/docs/api-reference/foundational-time-series-model-multi-series-historic.md): Based on the provided data, this endpoint predicts the in-sample period (historical period) values of multiple time series at once. It takes a JSON as an input containing information like the series frequency and historical data. (See below for a full description of the parameters.) The response con…
- [Foundational Time Series Model Online Multi Series Anomaly Detector](https://nixtla.io/docs/api-reference/foundational-time-series-model-online-multi-series-anomaly-detector.md): This endpoint performs online anomaly detection based on the provided data. It uses cross-validation for more robust detection of anomalies and it supports detection for univariate and multivariate scenarios. It takes a JSON as an input containing information like the series frequency and historical…
- [Get single Fine-tuned Model](https://nixtla.io/docs/api-reference/get-single-fine-tuned-model.md): Retrieve metadata for a previously fine-tuned model. The response contains the metadata of a model that you have fine-tuned and is available to make forecasts.
- [List Fine-tuned Models](https://nixtla.io/docs/api-reference/list-fine-tuned-models.md): List all the finetuned models that you have created. The response contains a list with the IDs of the models that you have fine-tuned and are available to make forecasts.
- [Validate Api Key](https://nixtla.io/docs/api-reference/validate-api-key.md)
- [Audit and Clean Data](https://nixtla.io/docs/data_requirements/audit_clean.md): Learn how to audit and clean your data with TimeGPT.
- [Data Requirements](https://nixtla.io/docs/data_requirements/data_requirements.md): Overview of the data format and requirements for TimeGPT forecasting.
- [Missing Values](https://nixtla.io/docs/data_requirements/missing_values.md): Learn how to handle missing values in time series data for accurate forecasting with TimeGPT.
- [Multiple Time Series](https://nixtla.io/docs/data_requirements/multiple_series.md): Learn how to handle missing values in time series data for accurate forecasting with TimeGPT.
- [Cross-validation Tutorial](https://nixtla.io/docs/forecasting/evaluation/cross_validation.md): Master time series cross-validation with TimeGPT. Complete Python tutorial for model validation, rolling-window techniques, and prediction intervals with code examples.
- [Evaluation Metrics](https://nixtla.io/docs/forecasting/evaluation/evaluation_metrics.md): Learn to select the right evaluation metrics to measure the performance of TimeGPT.
- [Evaluation Pipeline](https://nixtla.io/docs/forecasting/evaluation/evaluation_utilsforecast.md): Learn how to evaluate TimeGPT model performance using tools in utilforecast
- [Categorical Variables](https://nixtla.io/docs/forecasting/exogenous-variables/categorical_features.md): Learn how to incorporate external categorical variables in your TimeGPT forecasts to improve accuracy.
- [Date/Time Features](https://nixtla.io/docs/forecasting/exogenous-variables/date_features.md): Learn how to incorporate date/time features into your forecasts to improve performance.
- [Holidays & Special Dates](https://nixtla.io/docs/forecasting/exogenous-variables/holiday_and_special_dates.md): Guide to using holiday calendar variables and special dates to improve forecast accuracy in time series.
- [Model Interpretability](https://nixtla.io/docs/forecasting/exogenous-variables/interpretability_with_shap.md): Learn how to interpret model predictions using SHAP values to understand the impact of exogenous variables.
- [Numeric Variables](https://nixtla.io/docs/forecasting/exogenous-variables/numeric_features.md): Learn how to incorporate external numeric variables to improve your forecasting accuracy.
- [Fine-tuning with a Specific Loss Function](https://nixtla.io/docs/forecasting/fine-tuning/custom_loss.md): Learn how to fine-tune a model using specific loss functions, configure the Nixtla client, and evaluate performance improvements.
- [Controlling the Level of Fine-Tuning](https://nixtla.io/docs/forecasting/fine-tuning/depth.md): Learn how to use the finetune_depth parameter to control the extent of fine-tuning in TimeGPT models.
- [Re-using fine-tuned models](https://nixtla.io/docs/forecasting/fine-tuning/save_reuse_delete_finetuned_models.md): Learn how to save, fine-tune, list, and delete TimeGPT models to optimize forecasting.
- [Fine-tuning Tutorial TimeGPT](https://nixtla.io/docs/forecasting/fine-tuning/steps.md): Adapt TimeGPT to your specific datasets for more accurate forecasts
- [Distributed Forecasting with Spark, Dask & Ray](https://nixtla.io/docs/forecasting/forecasting-at-scale/computing_at_scale.md): Scale your time series forecasting with TimeGPT using Spark, Dask, or Ray. Learn distributed computing for millions of time series with Python code examples and best practices.
- [Time Series Forecasting with Dask](https://nixtla.io/docs/forecasting/forecasting-at-scale/dask.md): Scale pandas workflows with Dask and TimeGPT for distributed time series forecasting. Learn to process 10M+ time series in Python with minimal code changes.
- [Time Series Forecasting with Ray](https://nixtla.io/docs/forecasting/forecasting-at-scale/ray.md): Scale machine learning pipelines with Ray and TimeGPT for distributed time series forecasting. Learn to integrate TimeGPT with Ray for complex ML workflows in Python.
- [Time Series Forecasting with Spark](https://nixtla.io/docs/forecasting/forecasting-at-scale/spark.md): Scale enterprise time series forecasting with Spark and TimeGPT. Learn to process 100M+ observations across distributed clusters with Python and PySpark.
- [Improve Forecast Accuracy with TimeGPT](https://nixtla.io/docs/forecasting/improve_accuracy.md): Advanced techniques to enhance TimeGPT forecast accuracy for energy and electricity.
- [Long-Horizon Forecasting with TimeGPT](https://nixtla.io/docs/forecasting/model-version/longhorizon_model.md): Master long-horizon time series forecasting in Python using TimeGPT. Learn to predict 2+ seasonal periods ahead with confidence intervals and uncertainty quantification.
- [Uncertainty Quantification with TimeGPT](https://nixtla.io/docs/forecasting/probabilistic/introduction.md): Learn how to generate quantile forecasts and prediction intervals to capture uncertainty in your forecasts.
- [Prediction Intervals](https://nixtla.io/docs/forecasting/probabilistic/prediction_intervals.md): Learn how to create prediction intervals with TimeGPT
- [Quantile Forecasts](https://nixtla.io/docs/forecasting/probabilistic/quantiles.md): Learn how to generate quantile forecasts with TimeGPT
- [Bounded Forecasts](https://nixtla.io/docs/forecasting/special-topics/bounded_forecasts.md): Learn how to generate forecasts with upper and lower bounds to match your business constraints.
- [Hierarchical Forecasting](https://nixtla.io/docs/forecasting/special-topics/hierarchical_forecasting.md): Learn how to use TimeGPT for hierarchical forecasting across multiple levels.
- [Irregular Timestamps](https://nixtla.io/docs/forecasting/special-topics/irregular_timestamps.md): Learn how to work with both regular and irregular timestamps in TimeGPT for accurate forecasting.
- [Temporal Hierarchical Forecasting with TimeGPT](https://nixtla.io/docs/forecasting/special-topics/temporal_hierarchical.md): Learn how to combine forecasts at different time frequencies to improve accuracy.
- [Quickstart (TimeGPT-2)](https://nixtla.io/docs/forecasting/timegpt_2_family.md): Learn how to use TimeGPT-2 family of time series forecasting models
- [Quickstart (TimeGPT-1)](https://nixtla.io/docs/forecasting/timegpt_quickstart.md): Learn how to use TimeGPT for accurate time series forecasting
- [About TimeGPT](https://nixtla.io/docs/introduction/about_timegpt.md): Learn about TimeGPT - the foundation model for time series.
- [TimeGPT FAQ](https://nixtla.io/docs/introduction/faq.md): Frequently asked questions about TimeGPT
- [Introduction](https://nixtla.io/docs/introduction/introduction.md): Welcome to TimeGPT - The foundational model for time series forecasting and anomaly detection
- [TimeGPT Subscription Plans](https://nixtla.io/docs/introduction/timegpt_subscription_plans.md): Overview of TimeGPT's Enterprise subscription plans with deployment options, support, and trial details.
- [Why TimeGPT?](https://nixtla.io/docs/introduction/why_timegpt.md): Understand the benefits of using TimeGPT for time series analysis.
- [Date Features](https://nixtla.io/docs/reference/date_features.md): Use holidays flags and special dates to improve your accuracy
- [SDK Reference](https://nixtla.io/docs/reference/sdk_reference.md)
- [TimeGPT Excel Add-in (Beta)](https://nixtla.io/docs/reference/timegpt_excel_add_in_beta_.md): Use TimeGPT from Microsoft Excel
- [TimeGPT in R](https://nixtla.io/docs/reference/timegpt_in_r.md): Using TimeGPT for time series forecasting in the R programming language
- [TimeGEN-1 Quickstart (Azure)](https://nixtla.io/docs/setup/azureai.md): Quickstart guide to deploy and use TimeGEN-1 on Azure with the Nixtla Python SDK for time series forecasting.
- [Docker Image for TimeGPT](https://nixtla.io/docs/setup/docker.md): Learn how to access TimeGPT via a Docker image
- [Python Wheel for TimeGPT](https://nixtla.io/docs/setup/python_wheel.md): Learn how to access TimeGPT via a Python wheel
- [Setting up your API key](https://nixtla.io/docs/setup/setting_up_your_api_key.md): Learn how to securely configure your Nixtla SDK API key using direct code or environment variables.
- [Forecasting Bitcoin Prices](https://nixtla.io/docs/use_cases/bitcoin_price_prediction.md): Master Bitcoin price forecasting with TimeGPT. Complete Python tutorial covering cryptocurrency prediction, anomaly detection, uncertainty quantification, and risk management strategies.
- [Forecasting Energy Demand](https://nixtla.io/docs/use_cases/forecasting_energy_demand.md): Energy demand forecasting tutorial using TimeGPT AI. Step-by-step Python guide for electricity consumption prediction with 90% faster predictions and superior accuracy.
- [Forecasting Intermittent Demand](https://nixtla.io/docs/use_cases/forecasting_intermittent_demand.md): Master intermittent demand forecasting with TimeGPT for inventory optimization. Achieve 14% better accuracy than specialized models using the M5 dataset with exogenous variables and log transforms.
- [Forecasting Web Traffic](https://nixtla.io/docs/use_cases/forecasting_web_traffic.md): Learn how to predict website traffic patterns using TimeGPT.
- [Logging and Serving with MLFlow](https://nixtla.io/docs/use_cases/logging_and_serving_with_mlflow.md): Use MLFlow to log experiment metrics using TimeGPT and serve TimeGPT
- [What-If Forecasting: Price Effects in Retail](https://nixtla.io/docs/use_cases/what_if_forecasting_price_effects_in_retail.md): Master what-if forecasting with TimeGPT for retail pricing optimization. Learn scenario analysis to predict demand changes from price adjustments using the M5 dataset. Step-by-step Python tutorial.

## OpenAPI Specs

- [openapi](https://nixtla.io/docs/openapi.json)

## Optional

- [Home](https://www.nixtla.io)
- [Get in touch](https://nixtla.io/book-a-free-trial?utm_source=nixtla.io&utm_campaign=/docs/timegpt/getting-started)
- [Meet with us](https://nixtla.io/book-a-call?utm_source=nixtla.io&utm_campaign=/docs/timegpt/getting-started)