AI Tools.

Search

time series forecasting

chronos-2

AutoGluon's distribution of Amazon's Chronos-2 time-series foundation model, packaged for use within the AutoGluon machine learning framework. The model uses a T5 encoder-decoder over quantized time-series tokens for zero-shot forecasting across diverse domains. AutoGluon wraps it in a high-level API for automated time-series modeling pipelines.

Last reviewed

Use cases

  • Zero-shot forecasting within AutoGluon time-series pipelines
  • Automated model selection comparing Chronos-2 against classical baselines
  • Rapid prototyping via AutoGluon's high-level TimeSeriesPredictor API
  • Benchmarking foundation model vs. dataset-specific models in a unified framework
  • Enterprise AutoGluon deployments needing foundation model time-series forecasting

Pros

  • AutoGluon API simplifies usage without manual T5 tokenization code
  • Apache 2.0 license
  • Zero-shot domain transfer without per-dataset training
  • AutoGluon handles preprocessing, ensembling, and evaluation automatically

Cons

  • AutoGluon dependency adds significant package overhead for single-model use
  • Same accuracy limitations as amazon/chronos-2 — token quantization error persists
  • Less flexible for custom inference code than the amazon/chronos-2 checkpoint directly
  • AutoGluon's opinionated pipeline may not suit all production architectures
  • Performance no different from amazon/chronos-2 — just a different packaging

FAQ

What is chronos-2 used for?

Zero-shot forecasting within AutoGluon time-series pipelines. Automated model selection comparing Chronos-2 against classical baselines. Rapid prototyping via AutoGluon's high-level TimeSeriesPredictor API. Benchmarking foundation model vs. dataset-specific models in a unified framework. Enterprise AutoGluon deployments needing foundation model time-series forecasting.

Is chronos-2 free to use?

chronos-2 is an open-source model published on HuggingFace. License terms vary by model — check the model card for the specific license.

How do I run chronos-2 locally?

Most HuggingFace models can be loaded with transformers or the appropriate framework library. See the model card for framework-specific instructions and hardware requirements.

Tags

chronos-forecastingsafetensorst5time seriesforecastingfoundation modelspretrained modelstime-series-forecastingdataset:autogluon/chronos_datasetsdataset:Salesforce/GiftEvalPretrainarxiv:2403.07815arxiv:2510.15821license:apache-2.0region:us