66M Downloads

One engine to predict
sales

By the team behind the world's most-used forecasting tools.

One deployment. Every use case.

Our most advanced product for turning data into production-ready predictions. More than a foundation model, it is a complete and customizable solution based on TimeGPT, Nixtla’s OSS ecosystem, and agentic automation.

T0IngestData plane
T1EngineOne product surface
T2InsightConsumption layer

Data

Programmatically connect to the systems where your time series already live, from warehouses and lakehouses to files, dataframes, and streams.

DatabricksLAKEHOUSE
SnowflakeWAREHOUSE
S3 + GCS + ADLSOBJECT STORAGE
Pandas + Polars + SparkDATAFRAMES
Excel + CSVSPREADSHEETS

+ many more

Nixtla Engine

An intelligent system that automatically builds the right pipeline for each use case, accessible through a dedicated Python package or API.

50%+more accurate
Millionsof time series
Secondsto predict
3 linesto run the engine

/ CORE

Multi-domain Model Suite

One interface for 200+ forecasting models, including TimeGPT and models from the Nixtlaverse.

ForecastingMULTIVARIATE
Anomaly DetectionREAL-TIME
ExplainabilityDRIVER ANALYSIS
What-if ScenariosSCENARIO PLANNING

/tools

Practitioner Toolkit

Built on Nixtla's OSS ecosystem, a complete set of tools for building and serving production workflows.

Plotting & Analysis
Preprocessing functions
Evaluation & Diagnostics

/Agentic

Automation Layer

Connect any AI assistant to the Nixtla Engine through our MCP to automate forecasting workflows.

Data Processing & Analysis
Autonomous Research
Explainability

Insights

Turn forecasts into action by delivering predictions, alerts, and runbooks across the business surfaces your teams use every day.

Jupyter + Databricks + SnowflakeNOTEBOOKS
Tableau + Power BI + LookerBI DASHBOARDS
Embedded apps + RESTAPPLICATIONS
Slack + TeamsALERTS

+ many more

Security & Governance

Policy controls, audit trails, team isolation, and approvals across supported deployment layers.

SSO
RBAC
AUDIT LOG
SOC 2 Type II
VPC · BYOC

Real outcomes delivered to customers

[50%]

Increase in accuracy

Leading to increased revenue and reduced costs

[85%]

Increase in efficiency

Faster insights and less cost

[200%]

Gains in time to value

Act faster and more efficiently

5sDeployment Time

100BProcessed Tokens

Serving customers in over 60 regions worldwide, Microsoft supports a diverse client base that includes 95% of Fortune 500 companies, government bodies, and innovative startups.

85%Less False Positives

10xImprovement in Latency

Its dynamic pricing, optimized routing, and intelligent driver dispatch systems set industry benchmarks for technological innovation.

1,000+Financial Indicators Forecasted

No retrainingZero-Shot Deployment

Founded in 1875, Prudential offers life insurance, annuities, retirement planning, group insurance, and institutional investment management via its PGIM division.

10%Growth in Revenue

22%Reduction in Cloud Costs

Despite the war, Darnytsia demonstrated remarkable resilience by maintaining uninterrupted production through the 2022 Russian conflict while growing revenues by 32.8% in 2023. The company's state-of-the-art facilities can produce over 4 billion tablets and 400 million ampoules annually, alongside tens of millions of vials of antibiotics and infusions

2025NYSE Listing

$13BValuation

The platform emphasizes security, compliance, and transparency, and provides deep liquidity, automated market making, and innovative trading features tailored for institutional and professional traders.

80%Deployment Time Reduction

20%Forecast Accuracy Improvement

Intergrid is an energy technology company accelerating the shift from biomass-, gas-, and oil-based industrial heat to fully electrified systems. Through both asset optimization and a heat-as-a-service model, Intergrid manages forecasting, market participation, and operations for electric boilers and heat pumps across factories, greenhouses, and district heating networks, enabling customers to access emissions-free heat at predictable costs.

150%Efficiency in Prescriptions

+10%Forecast Accuracy

Serving over 650 organizations worldwide and operating from more than 30 global locations, Eversana achieved approximately $5 billion in annual revenue as of March 2025. This robust presence underscores its strategic influence in navigating complex healthcare landscapes and bringing innovative therapies to market

50%Cost Reduction

1 secForecast Workflow

Backed by Goldman Sachs Asset Management, GridStor is well-capitalized to pursue its mission of building a more resilient and sustainable energy infrastructure across the United States. In the past year, it expanded aggressively into ERCOT and CAISO regions, including a new 150 MW/300 MWh facility in Texas

2M+Time Series Forecasted

+15–18%Forecast Accuracy (MAE)

IMSS is responsible for the procurement, distribution, and management of medicines and medical supplies across a complex national network that includes regional administrative bodies (OOADs) and thousands of medical units, including hospitals and clinics.

40%Engagement Increase

1 weekRapid Implementation

Their product suite includes the Polludrone (ambient air quality monitor), Odosense (odour monitor), Dustroid (dust particulate monitor), AQBot (industrial gas monitor), Weathercom (meteorological station), and the cloud-based Envizom software platform

+365 hSaved

100%Uptime

The countertop IoT unit dispenses filtered tap water, enriched with natural flavor concentrates, and lets users customize each drink's flavor intensity, sweetness, and carbonation. More details can be found on their website, Re:Drink — Die Flaschenlose Getränkelösung

50%Reduction in Forecast Runtime

500KUnique IDs per Pipeline Run

Skulicity is a predictive analytics and forecasting platform built for demanding retail supply chains. They help businesses selling to major big-box retailers. Skulicity supports complex planning and replenishment needs across horticulture, perishable grocery, consumer packaged goods, and retail categories where demand volatility, short shelf life, and timing precision are critical

22,000+Daily Forecasts

ImprovedModel Performance

The platform uses Nixtla’s TimeGPT to forecast more than 500,000 forecasts each month.

34%Reduction in Forecast Error

365 daysForecast Horizon

Flieber is a multichannel inventory planning platform that helps retail brands streamline supply chain decisions. Its platform integrates sales, inventory, and supply chain data across multiple channels to support demand forecasting, inventory optimization, and replenishment planning.

Supply Chain

Predict the demand for your products and services at every SKU and location, accounting for promotions, holidays, and intermittent sales, to cut stockouts, excess inventory, and expedite costs.

Energy

Forecast node-level and market-wide prices, demand, and generation to optimize grid operations, battery dispatch, and trading decisions.

Financial Markets

Forecast financial markets prices, volumes, and correlations to optimize trading decisions.

Deployment Options

Deploy anywhere, your way

Choose the deployment option that best fits your infrastructure and security requirements.

"We are incredibly excited to partner with Nixtla. With TimeGEN-1, companies can deploy powerful time series forecasting models effortlessly, enhancing operational efficiency and accuracy while significantly reducing costs. This is a prime example of how Azure AI continues to drive innovation, making advanced AI accessible to every enterprise and developer"

Asha Sharma

Corporate Vice President, Azure AI

"Since integrating Nixtla's forecasting engine, our ML monitoring has transformed. We now experience an 85% drop in false positives and a 10x improvement in detection speed, directly reducing operational expenses and boosting ROI"

Anindya Saha

Staff Engineer, Machine Learning Platform

TimeGPT delivered highly accurate predictions across thousands of quarterly indicators out of the box.

John Oh

Director of Data Science, Prudential Financial

TimeGPT's forecasting accuracy and speed transformed our operations. We've reduced stockouts while simultaneously decreasing inventory costs – a winning combination we hadn't achieved with previous solutions like statistical models (e.g., Auto ARIMA) and ML algorithms like Facebook Prophet

Serhii Kolchyk

Head of Analytics and Digital Innovation

Nixtla's TimeGPT lets us focus on new markets and complex optimization strategies. It's a core enabler of how we scale industrial heat electrification quickly and safely.

Jaakko Hyppönen

Head of Analytics at Intergrid

Nixtla's seamless integration with our existing technology stack and BI tools has revolutionized our forecasting process. With TimeGPT, our team now makes faster, more confident, data-driven decisions that drive significant business value

Daniel Del Carpio

Data Scientist, Eversana

TimeGPT's onboarding is remarkably smooth and fast. Within seconds, we have actionable predictions ready to go. This is extremely impactful for our business. TimeGPT has been a game-changer for our budgeting process, operations, and board-level reporting. The speed, accuracy, and reliability have made it an essential part of our mission-critical workflows

Brett Rudder

Senior Manager of Market Analytics, GridStor

TimeGPT has transformed our forecasting capabilities, enabling near-instant, highly accurate predictions that empower proactive decision-making in environmental monitoring

Bhumik Nayak

Product Manager, Oizom

Implementing Nixtla's automated forecasting solution has transformed our operational efficiency. As a technical founder managing multiple business functions, I've been able to reallocate critical resources from manual prediction work to strategic initiatives that drive our core business growth and innovation

Alex Colucci

Co-founder & Tech Lead

"The biggest breakthrough for us was overcoming our deployment constraints without changing our environment. That flexibility made it possible to scale forecasting without rebuilding everything from the ground up"

Vinit Agrharkar

Data Engineer Intern

TimeGPT outperformed our legacy xgboost models, delivering more reliable forecasts with minimal tuning and scaling effortlessly to half-a-million forecasts per month.

Michael Carr

Chief, Quantitative Research

"The TimeGPT engine enabled us to build a much more powerful and accurate model that cut our runtime from five hours to under five minutes."

Fabricio Miranda

CEO, Flieber

Winning VN1 Competition with Nixtla AI

/PlaytheVideo+

Generate forecasts for the VN1 forecasting competition using Nixtla, specifically TimeGPT-1 and TimeGPT-1-long-horizon, and also the median ensemble of these two models...

Opus 4.5

Generate forecasts for the VN1 forecasting competition using Nixtla, specifically TimeGPT-1 and TimeGPT-1-long-horizon, and also the median ensemble of these two models...

Opus 4.5

Generate forecasts for the VN1 forecasting competition using Nixtla, specifically TimeGPT-1 and TimeGPT-1-long-horizon, and also the median ensemble of these two models...

Opus 4.5

Nixtla AI <> Cursor Integration

/PlaytheVideo+

1234567891011121314151617181920212223242526

"""
Train multiple Nixtla models and evaluate with MASE metrics
"""
import pandas as pd
import numpy as np
from functools import partial
from nixtla import NixtlaClient
from utilsforecast.evaluation import evaluate
from utilsforecast.losses import mase
 
# Initialize Nixtla client
nixtla_client = NixtlaClient()
 
# Load data
print("Loading data...")
train_df = pd.read_parquet('data/train_data.parquet')
test_df = pd.read_parquet('data/test_data.parquet')
 
# Convert ds to datetime if needed
train_df['ds'] = pd.to_datetime(train_df['ds'])
test_df['ds'] = pd.to_datetime(test_df['ds'])
 
print(f"Train data: {train_df.shape}")
print(f"Test data: {test_df.shape}")
print(f"Unique IDs in train: {train_df['unique_id'].nunique()}")
print(f"Unique IDs in test: {test_df['unique_id'].nunique()}")

Read the Nixtla Time Series Analysis Guide to train timegpt-1, timegpt-2, chronos2, tirex and timesfm-2.5 on the train_data.parquet. The frequency of the data is W-MON. Show me a matrix of the model with mase metrics with the test_data.parquet...

GPT-5.5

Secure and Compliant

Your data never leaves your own infrastructure. Zero data egress. Zero data retention. SOC 2 Type II.

Fine Tune on your data

A customized and self-improving model that learns from your data to get the best possible predictions.

Advanced Explainability

Transform your data into accurate, explainable, forecasts without the need to build and maintain complex models. Forecast Smarter. Act Faster.

Compliance

Complete Compliance & Security Readiness

Stay compliant with privacy and healthcare regulations. Our platform meets GDPR and HIPAA requirements, providing data protection and compliance monitoring for regulated industries.

[01]

Automated audit trails

Every action is logged and timestamped with immutable audit trails for complete regulatory compliance.

[02]

Compliance monitoring

Real-time monitoring ensures continuous compliance with industry standards and regulations.

A solution for everyone

Pricing Options

Build your own pipelines

Open Source libraries:

StatsForecast
MLForecast
NeuralForecast
HierarchicalForecast
UtilsForecast
CoreForecast

View on Github

Best-in-class forecasting for your business.

Everything in Open Source, plus:

Premium Models and Updates
Fine-tuned on your data
State-of-the-art accuracy
Scalability and performance
Enterprise-level security
SLAs and dedicated support

Book a Call

The Future of Forecasting is Autonomous

Free Trial Schedule a Demo

[INFO] Initializing Nixtla MCP Server…
[SUCCESS] ✓ Nixtla AI is ready to forecast
 
> User: Forecast my sales using state-of-the-art models
 
[THINKING] • Evaluating 250+ forecasting models
• Comparing TimeGPT, Chronos, and statistical methods
 
[SUCCESS] ✓ Forecast plan and results ready
└─ Best model: TimeGPT-2
└─ Accuracy: 94.2%
└─ Forecast saved to ./forecast.csv

One engine to predict
sales

Selected Enterprise Customers

One deployment. Every use case.