# Nixtla

> Nixtla builds TimeGPT, a foundation model for time series forecasting and anomaly detection, and maintains open source forecasting libraries for Python.

Agent instructions:
- Use the OpenAPI specification to discover supported website API operations and their authentication requirements.
- Use the API catalog to locate machine-readable resources before relying on page structure.
- Send TimeGPT inference requests from trusted server-side code, never directly from a browser.
- Treat API keys as secrets. Do not print them, commit them, place them in URLs, or include them in generated output.
- Interactive signup, email verification, API-key creation, and API-key rotation require a human in the Nixtla dashboard.
- Prefer canonical HTML URLs when citing Nixtla pages. For programmatic reading, use `/index.md` for the homepage, append `.md` to other public page URLs, or request `Accept: text/markdown`.

## Agent guide

- [Complete agent instructions](https://www.nixtla.io/agents.md)

## Documentation

- [Nixtla home](https://www.nixtla.io): Overview of TimeGPT, Nixtla products, and time series capabilities. [Markdown](https://www.nixtla.io/index.md).
- [Open source ecosystem](https://www.nixtla.io/ecosystem): Nixtla's open source libraries for statistical, machine learning, neural, and hierarchical forecasting. [Markdown](https://www.nixtla.io/ecosystem.md).
- [Industries](https://www.nixtla.io/industries): Time series applications across industries and operational workflows. [Markdown](https://www.nixtla.io/industries.md).
- [TimeGPT quickstart](https://www.nixtla.io/docs/forecasting/timegpt_quickstart): Build a first TimeGPT forecast with the Python SDK. [Markdown](https://www.nixtla.io/docs/forecasting/timegpt_quickstart.md).
- [Nixtla open source documentation](https://nixtlaverse.nixtla.io): Documentation for StatsForecast, NeuralForecast, MLForecast, HierarchicalForecast, and related libraries.
- [Nixtla blog](https://www.nixtla.io/blog): Technical articles, product announcements, and forecasting tutorials. [Markdown](https://www.nixtla.io/blog.md).
- [Nixtla brand kit](https://www.nixtla.io/brand-kit): Download Nixtla logos, fonts, colors, backgrounds, and design-system references. [Markdown](https://www.nixtla.io/brand-kit.md).
- [Nixtla GitHub organization](https://github.com/Nixtla): Source repositories, examples, releases, and issue trackers for Nixtla projects.

## API discovery

- [Website dashboard OpenAPI specification](https://www.nixtla.io/openapi.json)
- [API catalog](https://www.nixtla.io/.well-known/api-catalog)
- [TimeGPT inference OpenAPI specification](https://www.nixtla.io/docs/openapi.json)
- [TimeGPT REST API reference](https://www.nixtla.io/docs/api-reference)
- [Documentation index for agents](https://www.nixtla.io/docs/llms.txt)

## Primary links

- [Nixtla home](https://www.nixtla.io): Overview of TimeGPT, Nixtla products, and time series capabilities. [Markdown](https://www.nixtla.io/index.md).
- [Open source ecosystem](https://www.nixtla.io/ecosystem): Nixtla's open source libraries for statistical, machine learning, neural, and hierarchical forecasting. [Markdown](https://www.nixtla.io/ecosystem.md).
- [Industries](https://www.nixtla.io/industries): Time series applications across industries and operational workflows. [Markdown](https://www.nixtla.io/industries.md).
- [Success stories](https://www.nixtla.io/success-stories): Customer examples of forecasting and anomaly detection in production. [Markdown](https://www.nixtla.io/success-stories.md).
- [TimeGPT quickstart](https://www.nixtla.io/docs/forecasting/timegpt_quickstart): Build a first TimeGPT forecast with the Python SDK. [Markdown](https://www.nixtla.io/docs/forecasting/timegpt_quickstart.md).
- [Nixtla open source documentation](https://nixtlaverse.nixtla.io): Documentation for StatsForecast, NeuralForecast, MLForecast, HierarchicalForecast, and related libraries.
- [Nixtla blog](https://www.nixtla.io/blog): Technical articles, product announcements, and forecasting tutorials. [Markdown](https://www.nixtla.io/blog.md).
- [Nixtla brand kit](https://www.nixtla.io/brand-kit): Download Nixtla logos, fonts, colors, backgrounds, and design-system references. [Markdown](https://www.nixtla.io/brand-kit.md).
- [Nixtla GitHub organization](https://github.com/Nixtla): Source repositories, examples, releases, and issue trackers for Nixtla projects.
- [Company](https://www.nixtla.io/company): Information about Nixtla's mission and team. [Markdown](https://www.nixtla.io/company.md).
- [Researchers and students](https://www.nixtla.io/researchers-and-students): Resources and programs for time series researchers and students. [Markdown](https://www.nixtla.io/researchers-and-students.md).
- [Responsible disclosure](https://www.nixtla.io/bug-hunters): Report security issues responsibly for Nixtla, TimeGPT, and the open forecasting ecosystem. [Markdown](https://www.nixtla.io/bug-hunters.md).
- [Book a call](https://www.nixtla.io/book-a-call): Contact Nixtla about enterprise forecasting and deployment requirements. [Markdown](https://www.nixtla.io/book-a-call.md).
- [Book a free trial](https://www.nixtla.io/book-a-free-trial): Request a guided TimeGPT trial. [Markdown](https://www.nixtla.io/book-a-free-trial.md).
- [GenAI waitlist](https://www.nixtla.io/genai-waitlist): Request early access to Nixtla generative AI capabilities for time series analysis. [Markdown](https://www.nixtla.io/genai-waitlist.md).
- [TimeGPT-2.1 waitlist](https://www.nixtla.io/timegpt-2-waitlist): Request early access to TimeGPT-2.1 forecasting and anomaly-detection capabilities. [Markdown](https://www.nixtla.io/timegpt-2-waitlist.md).
- [Nixtla dashboard](https://dashboard.nixtla.io): Interactive account, workspace, usage, and API-key management.
- [Service status](https://www.nixtla.io/status): Current and historical status for Nixtla services. [Markdown](https://www.nixtla.io/status.md).

## Optional

- [Complete LLM reference](https://www.nixtla.io/llms-full.txt): Extended instructions, API quickstart, every primary page, and blog metadata.

## Complete API quickstart

1. Create a Nixtla account interactively at https://dashboard.nixtla.io.
2. In the dashboard, create an API key for the intended workspace and capture the secret when it is shown once.
3. Install the Python client and pandas with `pip install nixtla pandas`.
4. Store the secret in the `NIXTLA_API_KEY` environment variable.
5. Run this Python quickstart:

```python
import os
import pandas as pd
from nixtla import NixtlaClient

df = pd.DataFrame(
    {
        "unique_id": ["series-1"] * 12,
        "ds": pd.date_range("2025-01-01", periods=12, freq="MS"),
        "y": [10.0, 12.0, 13.0, 15.0, 16.0, 18.0, 19.0, 21.0, 22.0, 24.0, 25.0, 27.0],
    }
)

client = NixtlaClient(api_key=os.environ["NIXTLA_API_KEY"])
forecast = client.forecast(df=df, h=12, freq="MS")
print(forecast)
```
6. Validate the returned forecast before using it in an automated decision or production workflow.

## Every primary page

- [Nixtla home](https://www.nixtla.io): Overview of TimeGPT, Nixtla products, and time series capabilities. [Markdown](https://www.nixtla.io/index.md).
- [Open source ecosystem](https://www.nixtla.io/ecosystem): Nixtla's open source libraries for statistical, machine learning, neural, and hierarchical forecasting. [Markdown](https://www.nixtla.io/ecosystem.md).
- [Industries](https://www.nixtla.io/industries): Time series applications across industries and operational workflows. [Markdown](https://www.nixtla.io/industries.md).
- [Success stories](https://www.nixtla.io/success-stories): Customer examples of forecasting and anomaly detection in production. [Markdown](https://www.nixtla.io/success-stories.md).
- [TimeGPT quickstart](https://www.nixtla.io/docs/forecasting/timegpt_quickstart): Build a first TimeGPT forecast with the Python SDK. [Markdown](https://www.nixtla.io/docs/forecasting/timegpt_quickstart.md).
- [Nixtla open source documentation](https://nixtlaverse.nixtla.io): Documentation for StatsForecast, NeuralForecast, MLForecast, HierarchicalForecast, and related libraries.
- [Nixtla blog](https://www.nixtla.io/blog): Technical articles, product announcements, and forecasting tutorials. [Markdown](https://www.nixtla.io/blog.md).
- [Nixtla brand kit](https://www.nixtla.io/brand-kit): Download Nixtla logos, fonts, colors, backgrounds, and design-system references. [Markdown](https://www.nixtla.io/brand-kit.md).
- [Nixtla GitHub organization](https://github.com/Nixtla): Source repositories, examples, releases, and issue trackers for Nixtla projects.
- [Company](https://www.nixtla.io/company): Information about Nixtla's mission and team. [Markdown](https://www.nixtla.io/company.md).
- [Researchers and students](https://www.nixtla.io/researchers-and-students): Resources and programs for time series researchers and students. [Markdown](https://www.nixtla.io/researchers-and-students.md).
- [Responsible disclosure](https://www.nixtla.io/bug-hunters): Report security issues responsibly for Nixtla, TimeGPT, and the open forecasting ecosystem. [Markdown](https://www.nixtla.io/bug-hunters.md).
- [Book a call](https://www.nixtla.io/book-a-call): Contact Nixtla about enterprise forecasting and deployment requirements. [Markdown](https://www.nixtla.io/book-a-call.md).
- [Book a free trial](https://www.nixtla.io/book-a-free-trial): Request a guided TimeGPT trial. [Markdown](https://www.nixtla.io/book-a-free-trial.md).
- [GenAI waitlist](https://www.nixtla.io/genai-waitlist): Request early access to Nixtla generative AI capabilities for time series analysis. [Markdown](https://www.nixtla.io/genai-waitlist.md).
- [TimeGPT-2.1 waitlist](https://www.nixtla.io/timegpt-2-waitlist): Request early access to TimeGPT-2.1 forecasting and anomaly-detection capabilities. [Markdown](https://www.nixtla.io/timegpt-2-waitlist.md).
- [Nixtla dashboard](https://dashboard.nixtla.io): Interactive account, workspace, usage, and API-key management.
- [Service status](https://www.nixtla.io/status): Current and historical status for Nixtla services. [Markdown](https://www.nixtla.io/status.md).

## Blog

### Beyond Accuracy: What Makes a Good Forecast?

Accuracy is essential, but a good enterprise forecast must also support real decisions, use the right information, earn trust, and operate reliably at scale.

- Canonical HTML: https://www.nixtla.io/blog/beyond-accuracy
- Markdown alternate: https://www.nixtla.io/blog/beyond-accuracy.md
- Publication date: 2026-07-20
- Image URL: https://www.nixtla.io/blog/images/beyond-accuracy/editorial-forecast-closeup.webp
- Image alt: A jagged black time series passes through five translucent evaluation panels and exits as a rising green forecast arrow.

### TimeGPT Is Now on Microsoft Foundry: From Deploy to First Forecast

TimeGPT is now available through Microsoft Foundry, letting Azure customers deploy it inside their own subscription, billed as standard Azure consumption, without a separate vendor review.

- Canonical HTML: https://www.nixtla.io/blog/timegpt-azure-marketplace
- Markdown alternate: https://www.nixtla.io/blog/timegpt-azure-marketplace.md
- Publication date: 2026-07-15
- Image URL: https://www.nixtla.io/blog/images/timegpt-azure-marketplace/editorial-forecast-closeup.webp
- Image alt: A layered model and database stack connects to security controls, anomaly markers, and a green forecast.

### Still using Prophet or AutoARIMA for forecasting?

We benchmarked TimeGPT-2.1 against Databricks AutoML on electricity, weather, and retail demand datasets, comparing accuracy, runtime, and per-series win rates. TimeGPT-2.1 improved MAE, RMSE, MAPE, and WAPE across all three benchmarks.

- Canonical HTML: https://www.nixtla.io/blog/automl-vs-timegpt
- Markdown alternate: https://www.nixtla.io/blog/automl-vs-timegpt.md
- Publication date: 2026-05-28
- Image URL: https://www.nixtla.io/blog/images/automl-benchmark/editorial-forecast-closeup.webp
- Image alt: Three black time series converge through a cloud of points into one rising green forecast.

### Forecasting 145,000 Time Series in 7 Minutes with TimeGPT

TimeGPT forecasts all 145,063 Wikipedia web traffic series in 6.6 minutes, beating a top Kaggle statistical baseline by 18.8% on SMAPE, with zero feature engineering.

- Canonical HTML: https://www.nixtla.io/blog/wiki-traffic-forecast
- Markdown alternate: https://www.nixtla.io/blog/wiki-traffic-forecast.md
- Publication date: 2026-05-15
- Image URL: https://www.nixtla.io/blog/images/wiki-traffic-forecast/editorial-forecast-closeup.webp
- Image alt: Many small time series converge into one historical series and a rising green forecast.

### TimeGPT in Snowflake Just Got a Full Upgrade: Anomaly Detection, Explainability, and a One Command Install

The Nixtla Snowflake integration is now part of the official nixtla package. Run forecasting, anomaly detection, SHAP based explainability, and evaluation, all from pure SQL, all inside Snowflake.

- Canonical HTML: https://www.nixtla.io/blog/timegpt-snowflake-integration
- Markdown alternate: https://www.nixtla.io/blog/timegpt-snowflake-integration.md
- Publication date: 2026-02-23
- Image URL: https://www.nixtla.io/blog/images/timegpt-snowflake-integration/editorial-forecast-closeup.webp
- Image alt: A four-layer data block sends selected green streams into a forecast service and rising forecast line.

### Nixtla Enterprise Expands with Leading Foundation Models, MCP, and Agentic Capabilities

This release introduces three major capabilities that together expand Nixtla from a single-model offering into a full time series intelligence platform

- Canonical HTML: https://www.nixtla.io/blog/genai-announcement
- Markdown alternate: https://www.nixtla.io/blog/genai-announcement.md
- Publication date: 2025-12-18
- Image URL: https://www.nixtla.io/blog/images/genai-announcement/editorial-forecast-closeup.webp
- Image alt: Three time series enter stacked model cubes and flow through an agent loop to a green forecast.

### TimeGPT 2.1: The Next Generation of Foundation Models for Time Series Forecasting

Announcing the private preview of TimeGPT-2.1, the first multivariate model in the TimeGPT family.

- Canonical HTML: https://www.nixtla.io/blog/timegpt-2-1-announcement
- Markdown alternate: https://www.nixtla.io/blog/timegpt-2-1-announcement.md
- Publication date: 2025-12-10
- Image URL: https://www.nixtla.io/blog/images/timegpt-2-1-announcement/editorial-forecast-closeup.webp
- Image alt: Three black time series enter a translucent model cube and emerge as one green forecast.

### Supercharge Your Sales Forecasts: A Complete Guide to Exogenous Variables in MLForecast

Learn how to incorporate external factors like prices, promotions, and calendar patterns into your time series forecasts using MLForecast's exogenous variables.

- Canonical HTML: https://www.nixtla.io/blog/mlforecast-exogenous-variables
- Markdown alternate: https://www.nixtla.io/blog/mlforecast-exogenous-variables.md
- Publication date: 2025-12-05
- Image URL: https://www.nixtla.io/blog/images/mlforecast-exogenous-variables/editorial-forecast-closeup.webp
- Image alt: Calendar, price-tag, and currency inputs enter a feature grid that feeds a rising forecast.

### Forecasting Championship Results using Time Series and Nixtla

Learn how to forecast championship standings using Nixtla's StatsForecast library.

- Canonical HTML: https://www.nixtla.io/blog/championship-forecasting
- Markdown alternate: https://www.nixtla.io/blog/championship-forecasting.md
- Publication date: 2025-11-28
- Image URL: https://www.nixtla.io/blog/images/championship_forecasting/editorial-forecast-closeup.webp
- Image alt: Two time series feed a tournament bracket that selects a starred model and green forecast.

### Automatic Model Selection with StatsForecast for Time Series Forecasting

Stop testing statistical models manually. Use StatsForecast to automatically fit AutoARIMA, AutoETS, AutoCES, and AutoTheta models, then select the best performer for each series through cross-validation.

- Canonical HTML: https://www.nixtla.io/blog/statsforecast-automatic-model-selection
- Markdown alternate: https://www.nixtla.io/blog/statsforecast-automatic-model-selection.md
- Publication date: 2025-11-20
- Image URL: https://www.nixtla.io/blog/images/statsforecast-automatic-model-selection/editorial-forecast-closeup.webp
- Image alt: Five candidate time series pass through a model-selection grid, with one green forecast selected.

### Damage Detection in Engineering Structures Using Nixtla

Learn how to detect cracks and structural damage using Nixtla's Anomaly Detection pipeline while accounting for temperature-induced variations in sensor data.

- Canonical HTML: https://www.nixtla.io/blog/damagedetection
- Markdown alternate: https://www.nixtla.io/blog/damagedetection.md
- Publication date: 2025-10-15
- Image URL: https://www.nixtla.io/blog/images/damage_detection/editorial-forecast-closeup.webp
- Image alt: Sensors on a bridge connect to vibration traces, with one green trace highlighting a sharp disturbance.

### TimeGPT 2: The Next Generation of Foundation Models for Time Series Forecasting

Announcing the private preview of TimeGPT-2 Mini, TimeGPT-2, and TimeGPT-2 Pro, enterprise grade foundation models with up to 60% accuracy improvement, built for mission-critical time series forecasting.

- Canonical HTML: https://www.nixtla.io/blog/timegpt-2-announcement
- Markdown alternate: https://www.nixtla.io/blog/timegpt-2-announcement.md
- Publication date: 2025-10-14
- Image URL: https://www.nixtla.io/blog/images/timegpt-2-announcement/editorial-forecast-closeup.webp
- Image alt: Three distinct time series enter a translucent cube filled with data points and emerge as one green forecast.

### Anomaly Detection for Cloud Cost Monitoring with Nixtla

Learn how to build a synthetic cloud cost dataset and use Nixtla's algorithms to detect spikes, drifts, and level shifts. This approach helps teams monitor performance and prevent unexpected billing surprises.

- Canonical HTML: https://www.nixtla.io/blog/anomalyinproduction
- Markdown alternate: https://www.nixtla.io/blog/anomalyinproduction.md
- Publication date: 2025-09-26
- Image URL: https://www.nixtla.io/blog/images/anomaly_detection_monitoring/editorial-forecast-closeup.webp
- Image alt: A rising black time series with an extreme green spike beside connected storage blocks.

### Anomaly Detection in Time Series with TimeGPT and Python

Discover how to use TimeGPT for scalable, accurate anomaly detection in Python. Includes real-world time series, exogenous variables, and adjustable confidence levels.

- Canonical HTML: https://www.nixtla.io/blog/anomaly-detection
- Markdown alternate: https://www.nixtla.io/blog/anomaly-detection.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/anomaly_detection/editorial-forecast-closeup.webp
- Image alt: A rising black time series with four tall green spikes marked as detected anomalies.

### Performance Evaluation of Anomaly Detection through Synthetic Anomalies

Discover how to find the minimum detectable anomaly in absence of a ground truth labelled dataset using synthetic anomalies.

- Canonical HTML: https://www.nixtla.io/blog/anomaly-detection-performance-evaluation
- Markdown alternate: https://www.nixtla.io/blog/anomaly-detection-performance-evaluation.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/anomaly_detection_performance_evaluation/editorial-forecast-closeup.webp
- Image alt: A black time series crosses a shaded band and forms one tall green spike above the upper threshold.

### Automated Time Series Feature Engineering with MLforecast

Replace hours of custom feature engineering code with MLforecast's automated lag features, rolling statistics, and target transformations for faster, more reliable time series forecasting.

- Canonical HTML: https://www.nixtla.io/blog/automated-time-series-feature-engineering-with-mlforecast
- Markdown alternate: https://www.nixtla.io/blog/automated-time-series-feature-engineering-with-mlforecast.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/automated-time-series-feature-engineering-with-mlforecast/editorial-forecast-closeup.webp
- Image alt: A black time series enters a grid of engineered features and exits as a rising green forecast.

### Effortless Accuracy Unlocking the Power of Baseline Forecasts

Understand what are baseline forecasts, why they are important and learn to create them easily with Nixtla's statsforecast package.

- Canonical HTML: https://www.nixtla.io/blog/baseline-forecasts
- Markdown alternate: https://www.nixtla.io/blog/baseline-forecasts.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/baseline_forecasts/editorial-forecast-closeup.webp
- Image alt: One historical time series branches into rising, level, and falling baseline forecast paths.

### Eliminate Manual ARIMA Tuning Using StatsForecast AutoARIMA Automation

Eliminate weeks of manual ARIMA parameter tuning with StatsForecast's AutoARIMA. Automatically select optimal model parameters for 50+ time series with confidence intervals in under 30 minutes.

- Canonical HTML: https://www.nixtla.io/blog/eliminate-manual-arima-tuning-using-statsforecast-autoarima-automation
- Markdown alternate: https://www.nixtla.io/blog/eliminate-manual-arima-tuning-using-statsforecast-autoarima-automation.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/eliminate-manual-arima-tuning-using-statsforecast-autoarima-automation/editorial-forecast-closeup.webp
- Image alt: Five time series pass through toggle switches, with one green selection continuing as a forecast.

### Time Series Frequency Modelling with Fourier Transform and TimeGPT-1

Discover how to decompose your time series in multiple components with Fourier Transform and model each component with TimeGPT-1.

- Canonical HTML: https://www.nixtla.io/blog/frequency-ts-modelling
- Markdown alternate: https://www.nixtla.io/blog/frequency-ts-modelling.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/fourier_modelling/editorial-forecast-closeup.webp
- Image alt: A jagged time series is separated into three smooth frequency waves and recombined as a green forecast.

### Understanding Intermittent Demand

Learn how to forecast intermittent demand using Python and Nixtla's TimeGPT. This step-by-step guide covers handling sparse time series, fine-tuning, and using exogenous variables to improve accuracy.

- Canonical HTML: https://www.nixtla.io/blog/intermittent-demand
- Markdown alternate: https://www.nixtla.io/blog/intermittent-demand.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/intermittent_demand/editorial-forecast-closeup.webp
- Image alt: Sparse black demand spikes continue into a green forecast band with intermittent peaks.

### Long Term Mid Term and Short Term Forecasting with Polynomial Regression AutoARIMA and TimeGPT-1

Learn how to match forecasting models to your time horizon for better accuracy. Compare polynomial regression for long-term trends, AutoARIMA for mid-term cycles, and TimeGPT-1 for short-term predictions using real currency exchange data. Includes code examples for multi-horizon forecasting strategies.

- Canonical HTML: https://www.nixtla.io/blog/multilevelforecasting
- Markdown alternate: https://www.nixtla.io/blog/multilevelforecasting.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/MultiLevelForecasting/editorial-forecast-closeup.webp
- Image alt: Groups of small time series merge into three larger series and one green forecast line.

### Simple Anomaly Detection in Time Series via Optimal Baseline Subtraction (OBS)

Discover how to detect anomalies using Optimal Baseline Subtraction and enhance your forecasts with Nixtla’s TimeGPT on real-world weather data.

- Canonical HTML: https://www.nixtla.io/blog/optimal-baseline
- Markdown alternate: https://www.nixtla.io/blog/optimal-baseline.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/optimal_baseline/editorial-forecast-closeup.webp
- Image alt: A black time series is compared with a dotted purple baseline and green residual bars.

### Savitzky Golay Filtering for Time Series Denoising

Denoise your time series with polynomial smoothing using the Savitzky-Golay filter

- Canonical HTML: https://www.nixtla.io/blog/polynomial-filtering
- Markdown alternate: https://www.nixtla.io/blog/polynomial-filtering.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/Polynomial_Filtering/editorial-forecast-closeup.webp
- Image alt: A jagged black time series passes through a panel with a smooth green curve over it.

### Production-Ready Forecasting Pipeline with TimeGPT and Polars

Learn how TimeGPT's native DataFrame compatibility lets you leverage Polars' blazing-fast performance for time series forecasting without data conversion overhead.

- Canonical HTML: https://www.nixtla.io/blog/production-ready-forecasting-pipeline-with-timegpt-and-polars
- Markdown alternate: https://www.nixtla.io/blog/production-ready-forecasting-pipeline-with-timegpt-and-polars.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/production-ready-forecasting-pipeline-with-timegpt-and-polars/editorial-forecast-closeup.webp
- Image alt: Three table panels connect to a cube-shaped model that outputs a green forecast line.

### TimeGPT vs Snowflake - 50x Faster Forecasting with Better Accuracy

Discover SQL-native time series forecasting for Snowflake that's 50x faster than native tools. Nixtla provides state-of-the-art accuracy without Python, ML infrastructure, or complex setup.

- Canonical HTML: https://www.nixtla.io/blog/timegpt-in-snowflake
- Markdown alternate: https://www.nixtla.io/blog/timegpt-in-snowflake.md
- Publication date: 2025-08-26
- Image URL: https://www.nixtla.io/blog/images/timegpt_in_snowflake/editorial-forecast-closeup.webp
- Image alt: Four time series inputs connect to a central model cube that produces a green forecast with uncertainty bands.