Skip to main content
Kaino.dev
Discover
Evals
News
Academics
Insights
Kaino.dev

Discover, evaluate, and compare AI tools, models, and agents.

Explore

  • Discover
  • Evaluations
  • News
  • Academics
  • Insights

Community

  • Twitter
  • YouTube
  • Instagram
Privacy PolicyTerms of Service

© 2026 Kaino.dev. All rights reserved.

Version 1.1.0
autogluon/chronos-2 · Discover · Kaino
Discover/MODELS/autogluon/chronos-2
autogluon/chronos-2 logo

MODELS

autogluon/chronos-2

by AutoGluon

modellead-sourcehugging-face-popular-modelssource:github.comtime-series-forecastingchronos-2autogluonamazon-sciencehugging-face
Visit WebsiteDocumentationGitHub

Overview

Chronos-2 is a time-series forecasting model available on Hugging Face and documented for use with AutoGluon TimeSeriesPredictor.

Details

Chronos-2 appears as autogluon/chronos-2 on Hugging Face for time-series forecasting. AutoGluon documentation covers using Chronos-2 through TimeSeriesPredictor, including zero-shot forecasting, covariates, fine-tuning, LoRA defaults, and the chronos2_ensemble preset. The official amazon-science/chronos-forecasting repository lists Chronos-2 as released on October 20, 2025, provides model IDs including amazon/chronos-2, and includes pip installation plus Chronos2Pipeline inference usage.

When to Use

Use Chronos-2 when you need a time-series forecasting model that is documented for AutoGluon TimeSeriesPredictor workflows. Use it for zero-shot forecasting workflows or experiments involving covariates fine-tuning LoRA defaults or the chronos2_ensemble preset as described in the AutoGluon documentation. Use the official GitHub repository when you need installation and Chronos2Pipeline inference examples.

Getting Started

  1. Open the AutoGluon Chronos-2 forecasting documentation and review the TimeSeriesPredictor examples.
  2. Install from the official amazon-science/chronos-forecasting repository instructions
  3. which include pip installation guidance.
  4. Try the documented Chronos2Pipeline inference usage or the AutoGluon TimeSeriesPredictor workflow on a small time-series dataset.
  5. Compare the Hugging Face model page for autogluon/chronos-2 with the repository model IDs such as amazon/chronos-2 before standardizing a deployment reference.

Key Features

  • •Time-series forecasting model listed on Hugging Face.
  • •Documented integration with AutoGluon TimeSeriesPredictor.
  • •Supports zero-shot forecasting workflows according to AutoGluon documentation.
  • •AutoGluon documentation covers covariates
  • •fine-tuning
  • •LoRA defaults
  • •and the chronos2_ensemble preset.
  • •Official repository provides pip installation and Chronos2Pipeline inference usage.

Capabilities

  • •time-series-forecasting
  • •zero-shot-forecasting
  • •covariate-aware-forecasting
  • •fine-tuning
  • •autogluon-timeseriespredictor
  • •chronos2pipeline-inference

Last updated Jun 4, 2026