Nixtla
Nixtla is a private AI company building TimeGPT, a proprietary foundation model for time series forecasting and anomaly detection, serving enterprises across retail, energy, finance, healthcare, and mobility via an open-source ecosystem and a commercial platform.
- Company typePrivate
- Founded2022
- HeadquartersSan Francisco, United States
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
What Nixtla does
Nixtla is a private, founder-led AI company (founded circa 2022, headquartered in San Francisco, Delaware-incorporated) that builds TimeGPT, a proprietary generative pre-trained transformer foundation model purpose-built for time series forecasting and anomaly detection. TimeGPT is described as trained on the largest collection of time series data in history — over 100 billion rows spanning financial, weather, energy, and web data — and is held as a closed-source model with a family of variants (timegpt-1, timegpt-1-long-horizon, timegpt-2, timegpt-2-mini, plus azureai/TimeGEN-1 on Azure). The model is delivered via a REST API (api.nixtla.io), Python and R SDKs, a Microsoft AppSource Excel add-in, MLflow pyfunc serving, and a Model Context Protocol (MCP) agentic automation layer that orchestrates forecasting, anomaly detection, what-if scenario planning, and explainability. Wrapping TimeGPT is the Nixtlaverse, the largest open-source time series ecosystem — StatsForecast, MLForecast, NeuralForecast, HierarchicalForecast, UtilsForecast, CoreForecast, and DatasetsForecast — distributed under permissive licenses and reporting 66M+ downloads and 16k GitHub stars.
Nixtla operates a hybrid go-to-market: a free, permanent open-source tier feeds a self-serve PLG funnel and developer community (Slack, GitHub, Twitter/X); a quote-based TimeGPT Enterprise subscription adds fine-tuning on customer data, 200+ model access, SOC 2 Type II / GDPR / HIPAA-ready security (SSO, RBAC, VPC/BYOC), SLAs, and dedicated support; and a usage-based API meters input/output/finetune tokens. Distribution layers enterprise direct sales (Book a Call), hyperscaler marketplaces (Microsoft Azure AI for TimeGEN-1, Microsoft AppSource for the Excel add-in), data-platform channels (Databricks lakehouse, Snowflake warehouse, AWS/GCS/ADLS object storage), and an OEM/embedded partnership with Zone & Co that places TimeGPT inside an ERP-native financial operations platform. The company reports SOC 2 Type II, GDPR, and HIPAA compliance readiness, a public status page at 99.97% API uptime over the trailing 90 days, and customer-stated outcomes of up to 50% higher forecast accuracy, 85% efficiency gains, and 200% time-to-value improvement versus traditional methods, with 42% accuracy gains and 10x inference efficiency cited against legacy approaches.
Nixtla serves enterprise and government customers across at least eight verticals: supply chain and retail/CPG (Nestlé, Unilever, Zalando, Decathlon, Amazon, Walmart, Flieber, Skulicity); energy (GridStor, Intergrid); financial services and crypto (Prudential, TradeSmith, Bullish); healthcare and life sciences (IMSS, Darnytsia, Eversana); mobility (Lyft, Grab); environmental monitoring (Oizom); food and beverage (Re:Drink); and a hyperscaler channel through Microsoft Azure AI. Total disclosed funding is roughly $48M across six rounds through February 2026, most recently a $16M Series A led by Energize Capital with True Ventures and GreatPoint Ventures.
Nixtla firmographics
Firmographics- Name
- Nixtla
- Legal name
- Nixtla, Inc.
- Website
- https://nixtla.io
- Company type
- Private
- Founded year
- 2022
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- Nixtla is a private AI company building TimeGPT, a proprietary foundation model for time series forecasting and anomaly detection, serving enterprises across retail, energy, finance, healthcare, and mobility via an open-source ecosystem and a commercial platform.
- Ownership category
- akta.pro rank
Nixtla industry classification
Industry- Product category
- Time Series Forecasting Software
- NAICS
- Software Publishers (5132), Computer Systems Design and Related Services (5415), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182)
- SIC
- Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370)
- akta.pro primary industry
- End-to-End MLOps & ML Platform Suites (HDAAABAA)
- akta.pro secondary industry
- Pipeline Orchestration & Workflow Automation (HDAAABAD)
Keywords
Where Nixtla is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Offices1 record
Markets served
Nixtla business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Personnel, Marketing or Sales, Infrastructure, Operations
Revenue model
- Open Source / Free Tier: Permanently free open-source libraries (StatsForecast, MLForecast, NeuralForecast, HierarchicalForecast, UtilsForecast, CoreForecast) distributed via GitHub under permissive licenses; serves as a top-of-funnel acquisition channel rather than a direct revenue stream.
- TimeGPT Enterprise Subscription: Subscription-based Enterprise tier offering premium models and updates, fine-tuning on customer data, state-of-the-art accuracy, scalability/performance, enterprise security (SSO, RBAC, SOC 2 Type II, VPC, BYOC), SLAs, and dedicated support; pricing is quote-based via 'Book a Call'.
- TimeGPT API consumption: Token-based consumption of the TimeGPT REST API (api.nixtla.io) for /v2/forecast, /v2/cross_validation, /v2/anomaly_detection, /v2/online_anomaly_detection, /v2/finetune; usage-based billing reported via /usage endpoint with input/output/finetune tokens metered.
- TimeGEN-1 on Azure (consumption): TimeGEN-1, TimeGPT optimized for Microsoft Azure AI, available via dedicated Azure AI setup path; billed through Azure consumption alongside Nixtla's per-token fee schedule on the pricing page.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Pay-as-you-go | Free tier: open-source libraries for self-built pipelines (StatsForecast, MLForecast, NeuralForecast, HierarchicalForecast, UtilsForecast, CoreForecast). |
| Subscription | Annual | Enterprise tier: best-in-class forecasting with premium models, fine-tuning, scalability, enterprise security, SLAs, and dedicated support. |
| Usage-based | Pay-as-you-go | TimeGPT API: usage-based metered access at api.nixtla.io. |
Go-to-market motion5 records
Distribution channels8 records
Marketing channels11 records
Nixtla product offering
Product offeringCore offering
Nixtla develops and sells TimeGPT, a proprietary generative pre-trained transformer foundation model purpose-built for time series forecasting and anomaly detection, exposed via a REST API and Python/R SDKs. It also operates the Nixtlaverse, a suite of free open-source forecasting libraries (StatsForecast, MLForecast, NeuralForecast, HierarchicalForecast, UtilsForecast, CoreForecast), and the Nixtla Engine/Enterprise platform that unifies 200+ models, data-plane integrations (Databricks, Snowflake, S3, GCS, ADLS), and agentic MCP automation for enterprise customers.
Differentiator
Problem solved
Functional benefit
Brands
- TimeGPT: Nixtla's flagship foundation model for time series forecasting and anomaly detection, the basis of its enterprise TimeGPT-1 and TimeGPT-2 offerings.
- Nixtlaverse
- Nixtla Enterprise
Products and services
- TimeGPT Nixtla's flagship generative pre-trained transformer foundation model for time series forecasting and anomaly detection, trained on over 100 billion rows of data, exposed via REST API, Python SDK (NixtlaClient), R SDK, Excel add-in, and MCP server for zero-shot and fine-tuned forecasting.
- Nixtla Engine Unified deployment product turning data into production-ready predictions via three layers: Data plane (Databricks, Snowflake, S3/GCS/ADLS, dataframes, spreadsheets), Engine (Multi-domain Model Suite with 200+ models including TimeGPT, Practitioner Toolkit, Automation Layer via MCP), and Insights (notebooks, BI dashboards, embedded apps, Slack/Teams alerts).
- Nixtla Enterprise Commercial subscription tier offering premium models and updates, fine-tuning on customer data, scalability, enterprise security (SOC 2 Type II, SSO, RBAC, audit log, VPC/BYOC), SLAs, and dedicated support, sold via 'Book a Call' quote-based pricing.
- TimeGEN-1 TimeGPT optimized for Microsoft Azure AI deployment, enabling Azure customers to deploy Nixtla's time series forecasting models on Azure infrastructure through one-click deployment.
- StatsForecast Open-source Python library providing lightning-fast forecasting with statistical and econometric models including AutoARIMA, AutoETS, AutoCES, AutoTheta, MSTL, GARCH, ARCH, Holt-Winters, and Croston variants, with distributed computing via Dask, Ray, and Spark.
- MLForecast Open-source library providing scalable machine learning for time series forecasting, part of the Nixtlaverse ecosystem.
- NeuralForecast Open-source library providing neural network models for time series forecasting, part of the Nixtlaverse ecosystem.
- HierarchicalForecast Open-source library for hierarchical time series forecasting and reconciliation, part of the Nixtlaverse ecosystem.
- TimeGPT Excel Add-in Beta add-in distributed via Microsoft AppSource that enables TimeGPT forecasting and anomaly detection directly within Microsoft Excel using an API key, supporting configurable frequency and horizon.
- Bug Hunters Responsible Disclosure Program Open responsible-disclosure security research program offering rewards up to $10,000 (with P0 24-hour triage) for vulnerability reports against the TimeGPT API, Platform, nixtla.io, and open-source surfaces.
Quantifiable outcome
- Up to 42% more accurate forecasts and 10x inference efficiency vs traditional methods
- +13 more outcomes
Companies that use Nixtla
Customer profileNamed customers21 records
Segments7 records
Ideal customer profiles4 records
Nixtla technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration14 records
AI capability9 records
Feature11 records
Nixtla partnerships and signals
Strategic signalPartnerships
Nine partnerships are on record, tiered core, flagship and minor.
- Zone & CocoreStrategic partnership to embed TimeGPT, a foundation model for time series data, directly into Zone's ERP-native financial operations platform. Delivers continuous anomaly detection on billing/usage activity and foundation-model-powered cash flow forecasting without traditional data-science infrastructure; customer data governance ensures data is not shared for model training.
- MicrosoftflagshipDeep strategic and technology partnership: Microsoft hosts TimeGEN-1 (TimeGPT optimized for Azure) on Azure AI, enabling 'powerful time series forecasting models effortlessly' per Asha Sharma (Corporate VP, Azure AI). Microsoft serves customers in 60+ regions using Nixtla with 5-second deployment and 100B processed tokens.
- DatabrickscoreTechnology integration with Databricks lakehouse as a first-class Data plane (alongside Snowflake, S3/GCS/ADLS). Databricks-hosted tutorial introduces TimeGPT alongside Chronos, TimesFM, and Moirai with downloadable notebooks demonstrating TimeGPT forecasting in Databricks environments.
- SnowflakecoreTechnology integration with Snowflake as a first-class Data plane (warehouse). Nixtla Enterprise connects programmatically to Snowflake alongside Databricks, S3, GCS, ADLS, Pandas, Polars, and Spark.
- Amazon Web Services (S3 / GCS / ADLS object storage)coreTechnology integration with AWS S3 (plus Google Cloud Storage and Azure Data Lake Storage) as object storage data sources, plus one-click self-hosted deployment via Docker/Pip/Terraform on AWS infrastructure.
- MLflowcoreFirst-class integration with the MLflow open-source ML platform for experiment tracking (forecast, cross-validation, anomaly detection) and for serving TimeGPT as a unified mlflow.pyfunc.PythonModel with operational modes 'forecast', 'cross_validation', and 'anomaly_detection'.
- Microsoft AppSourcecoreDistribution partnership via Microsoft AppSource for the TimeGPT Excel Add-in (Beta), enabling finance and analyst users to run TimeGPT forecasting and anomaly detection directly inside Microsoft Excel.
- Amazon (Chronos)minorAdjacent technology relationship: Nixtla benchmarks TimeGPT alongside Amazon's Chronos foundation model in third-party Databricks coverage, positioning TimeGPT in the same category of foundation time-series models.
- Google Research (TimesFM)minorAdjacent technology relationship: Nixtla benchmarks TimeGPT alongside Google Research's TimesFM foundation model in third-party Databricks coverage.
Scale indicators14 records
Recent moves3 records
Expansion highlights6 records
Nixtla competitors and assessment
Company assessmentBroad incumbents
- Microsoft Azure AI (TimeGEN-1 host & Azure OpenAI competitor): Microsoft hosts TimeGEN-1 as an Azure AI service and is also Nixtla's flagship customer, but Azure AI increasingly bundles competing forecasting and anomaly detection capabilities. Directly comparable as a channel partner and as the largest incumbent in the AI/ML platform category.
- AWS Forecast: Amazon's managed time-series forecasting service on AWS, directly competing with Nixtla's TimeGPT for cloud-native demand forecasting, capacity planning, and financial planning workloads. Both target enterprise developers with API-first forecasting, but AWS bundles forecasting inside the broader AWS ecosystem rather than specializing.
- Google Vertex AI Forecasting (TimesFM): Google's Vertex AI offers time-series forecasting plus its open-source TimesFM foundation model, which Nixtla explicitly benchmarks against TimeGPT. Comparable in offering foundation-model-driven forecasting, anomaly detection, and enterprise MLOps tooling, but bundled into Google's broader AI/ML platform.
- Salesforce Einstein / Moirai: Salesforce's predictive analytics and forecasting suite built around the open-source Moirai foundation model for time series. Comparable to Nixtla in delivering zero-shot forecasting at enterprise scale, but embedded in the Salesforce CRM/cloud ecosystem rather than offered as a standalone time-series intelligence layer.
- Anomaly.io (Datadog Watchdog / Splunk ITSI): Datadog and Splunk offer anomaly detection and AIOps on time-series telemetry/metrics at scale, overlapping with Nixtla's online anomaly detection capabilities (/v2/online_anomaly_detection). Comparable in serving ML-monitoring and operational-anomaly use cases, but specialized in observability/IT data rather than cross-domain business forecasting.
- Databricks (built-in forecasting & lakehouse AI): Databricks is both a Nixtla technology partner (lakehouse integration, Databricks-hosted TimeGPT tutorials) and a competitive incumbent offering its own time-series forecasting, MLflow, and Mosaic AI stack. Directly comparable for enterprise data teams running forecasting on lakehouse data.
- Snowflake (built-in forecasting & Cortex): Snowflake is both a Nixtla integration partner (warehouse data plane) and a competitive incumbent with native forecasting functions and Snowflake Cortex AI. Comparable for enterprise customers running forecasting workloads directly inside the data warehouse without a separate vendor.
Emerging players
- MindsDB: Open-source AI/ML platform that brings ML models — including time-series forecasting and anomaly detection — directly into databases and data warehouses. Comparable to Nixtla in open-source-led enterprise positioning and ML-on-data-plane model, but broader in scope (NLP, classification, regression) and less specialized in time-series foundation models.
Direct peers
- DataRobot: Enterprise AI/ML platform offering automated forecasting, anomaly detection, and predictive use cases. Closely comparable to Nixtla Enterprise in target buyer (enterprise data teams), delivery model (managed platform), and use cases (time-series + broader predictive analytics), though DataRobot is a broader AutoML suite rather than a time-series specialist.
- H2O.ai: AI cloud and foundation-model platform providing automated machine learning, time-series forecasting, and predictive analytics for enterprise customers. Comparable to Nixtla in combining open-source libraries (H2O, Driverless AI) with an enterprise platform and in targeting forecasting and anomaly detection use cases.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat6 records
Key risks6 records
Key highlights7 records
Customer concentration
Nixtla social profiles
Digital presenceNixtla compliance and trust
Trust signalCompliance3 records
Nixtla financial estimates
Financial estimateRevenue estimate
Valuation estimate
Nixtla leadership team
Management profileNumber of profiles
Profiles2 records
Nixtla funding detail
Funding detailFunding overview
Funding rounds6 records
Investors10 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Nixtla M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Nixtla
What does Nixtla do?
Nixtla develops and sells TimeGPT, a proprietary generative pre-trained transformer foundation model purpose-built for time series forecasting and anomaly detection, exposed via a REST API and Python/R SDKs. It also operates the Nixtlaverse, a suite of free open-source forecasting libraries (StatsForecast, MLForecast, NeuralForecast, HierarchicalForecast, UtilsForecast, CoreForecast), and the Nixtla Engine/Enterprise platform that unifies 200+ models, data-plane integrations (Databricks, Snowflake, S3, GCS, ADLS), and agentic MCP automation for enterprise customers.
Is Nixtla a public or private company?
Nixtla is a private company. It is classified as venture growth investor backed and is currently operating.
When was Nixtla founded?
Nixtla was founded in 2022. It employs 11 to 50 people.
Where is Nixtla based?
Nixtla is headquartered in San Francisco, United States, in the North America region.
How does Nixtla make money?
Four revenue lines are on record. Open Source / Free Tier is the primary driver. The others are timeGPT Enterprise Subscription, timeGPT API consumption and timeGEN-1 on Azure (consumption).
Who are Nixtla's main competitors?
Broad incumbents on record are Microsoft Azure AI (TimeGEN-1 host & Azure OpenAI competitor), AWS Forecast, Google Vertex AI Forecasting (TimesFM), Salesforce Einstein / Moirai, Anomaly.io (Datadog Watchdog / Splunk ITSI), Databricks (built-in forecasting & lakehouse AI) and Snowflake (built-in forecasting & Cortex). MindsDB is listed as an emerging player. Direct peers are DataRobot and H2O.ai.
Does Nixtla have an API?
Yes. Public REST API at https://api.nixtla.io providing access to TimeGPT foundation model endpoints including forecasting (POST /v2/forecast), cross-validation (POST /v2/cross_validation), online and historical anomaly detection (POST /v2/online_anomaly_detection and POST /v2/anomaly_detection), fine-tuning (POST /v2/finetune), fine-tuned model management (GET/DELETE /v2/finetuned_models), and API key validation (GET /validate_api_key). Authentication uses Bearer token. Model options include timegpt-1, timegpt-1-long-horizon, timegpt-2, timegpt-2-mini, and azureai (TimeGEN-1 for Azure). Also offered as an Azure AI deployment (TimeGEN-1). Available SDKs: Python (NixtlaClient) and R (nixtlar on CRAN). Developer documentation is at www.nixtla.io/docs.
What industry is Nixtla in?
Nixtla's product category is Time Series Forecasting Software. Its primary akta.pro industry code is HDAAABAA, End-to-End MLOps & ML Platform Suites, with a secondary code of HDAAABAD, Pipeline Orchestration & Workflow Automation. Its NAICS code is 5132 and its SIC code is 7372.