Update Jun 5, 2026: Deploy Chronos-2 on AWS with AutoGluon-Cloud. Real-time, serverless, or batch inference in 3 lines of code — pandas DataFrames in, forecasts out. Check out the new deployment guide. Chronos-2 is a 120M-parameter, encoder-only time series foundation model for zero-shot forecasting. It supports univariate, multivariate, and covariate-informed tasks within a single architecture. Inspired by the T5 encoder, Chronos-2 produces multi-step-ahead quantile forecasts and uses a group attention mechanism for efficient in-context learning across related series and covariates. Trained on a combination of real-world and large-scale synthetic datasets, it achieves state-of-the-art…
This is a variant of the Chronos-2 model which has only been trained on synthetic univariate and multivariate data. For usage and details on the Chronos-2 model, please refer to https://huggingface.co/autogluon/chronos-2.
Parameters119M
Context—
Weights476.0 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads470.2k
Runs On
What it takes to serve chronos-2-synth (119M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.2 GB | 0.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
SAVRN's Notes on chronos-2-synth
The word that matters in this name is synth. Autogluon trained this 119-million-parameter Chronos-2 variant only on synthetic univariate and multivariate series. Memory is not the decision point. At 16-bit the weights are 0.2 GB and it runs in 0.3 GB, and the single 192 GB MI300X that heads our table at $1.85 per hour on-demand would sit mostly empty. Give it a slice of a card that also hosts the models the forecasts feed into.
The license is Apache 2.0, which permits commercial use, modification and redistribution with the notices kept and changes stated, and the weights are open access in safetensors. Follow the publisher's pointer: usage details live with the main Chronos-2 model, and arXiv:2510.15821 describes the method. Our record carries no context length figure, so test the history window on your own series, and decide what synthetic-only training means before it reaches production.
Model Card
By Autogluon, published under apache-2.0, revision 3607918a9fd0.
This is a variant of the Chronos-2 model which has only been trained on synthetic univariate and multivariate data. For usage and details on the Chronos-2 model, please refer to https://huggingface.co/autogluon/chronos-2. If you find Chronos-2 useful for your research, please consider citing the associated paper
Read Autogluon's full model card
This is a variant of the Chronos-2 model which has only been trained on synthetic univariate and multivariate data. For usage and details on the Chronos-2 model, please refer to https://huggingface.co/autogluon/chronos-2.
Citation
If you find Chronos-2 useful for your research, please consider citing the associated paper:
@article{ansari2025chronos2,
title = {Chronos-2: From Univariate to Universal Forecasting},
author = {Abdul Fatir Ansari and Oleksandr Shchur and Jaris Küken and Andreas Auer and Boran Han and Pedro Mercado and Syama Sundar Rangapuram and Huibin Shen and Lorenzo Stella and Xiyuan Zhang and Mononito Goswami and Shubham Kapoor and Danielle C. Maddix and Pablo Guerron and Tony Hu and Junming Yin and Nick Erickson and Prateek Mutalik Desai and Hao Wang and Huzefa Rangwala and George Karypis and Yuyang Wang and Michael Bohlke-Schneider},
year = {2025},
url = {https://arxiv.org/abs/2510.15821}
}
Configuration
- Architecture
- Chronos2Model
- Vocabulary size
- 2
- RoPE base
- 10000
- Stored precision
- float32
- Model type
- t5
Identity and Version
- Repository
- autogluon/chronos-2-synth
- Publisher
- Autogluon
- Task
- Time series forecasting
- Modality
- Time series
- Library
- chronos-forecasting
- Parameters
- 119M parameters
- Languages
- Not stated by the source
- Revision
- 3607918a9fd027d5c465d8213e46b98e2c041cea
- First published
- 2025-11-24
- Last updated
- 2025-12-03
Files and Weights
4 files, 476.0 MB in total. The weights are 1 file totalling 476.0 MB in safetensors.
Weights1 file · 476.0 MB
Configuration1 file · 971 B
Documentation1 file · 1.3 KB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 476.0 MB | 920a3344726a |
| config.json | Configuration | 971 B | — |
| README.md | Documentation | 1.3 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 476.0 MB
Download from Autogluon
Released by Autogluon through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:2510.15821
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 476.0 MB |
| 16-bit | 0.2 GB |
| 8-bit | 0.1 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Compare chronos-2-synth
Questions About chronos-2-synth
How much GPU memory does chronos-2-synth need?
About 0.3 GB at 16-bit and 0.1 GB at 4-bit: the weights (119M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run chronos-2-synth on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use chronos-2-synth commercially?
Yes. chronos-2-synth is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Similar Models
Update Jun 5, 2026: Deploy Chronos-2 on AWS with AutoGluon-Cloud. Real-time, serverless, or batch inference in 3 lines of code — pandas DataFrames in, forecasts out. Check out the new deployment guide. Chronos-2 is a 120M-parameter, encoder-only time series foundation model for zero-shot forecasting. It supports univariate, multivariate, and covariate-informed tasks within a single architecture. Inspired by the T5 encoder, Chronos-2 produces multi-step-ahead quantile forecasts and uses a group attention mechanism for efficient in-context learning across related series and covariates. Trained on a combination of real-world and large-scale synthetic datasets, it achieves state-of-the-art…
This model has been pushed to the Hub using the PytorchModelHubMixin integration: This release is for research purposes only in support of an academic paper. Our models, datasets, and code are not specifically designed or evaluated for all downstream purposes. We strongly recommend users evaluate and address potential concerns related to accuracy, safety, and fairness before deploying this model. We encourage users to consider the common limitations of AI, comply with applicable laws, and leverage best practices when selecting use cases, particularly for high-risk scenarios where errors or misuse could significantly impact people’s lives, rights, or safety. For further guidance on use…
MOMENT is a family of foundation models for general-purpose time-series analysis. The models in this family (1) serve as a building block for diverse time-series analysis tasks (e.g., forecasting, classification, anomaly detection, and imputation, etc.), (2) are effective out-of-the-box, i.e., with no (or few) task-specific exemplars (enabling e.g., zero-shot forecasting, few-shot classification, etc.), and (3) are tunable using in-distribution and task-specific data to improve performance. For details on MOMENT models, training data, and experimental results, please refer to the paper MOMENT: A Family of Open Time-series Foundation Models. Recommended Python Version: Python 3.11 (support…
This repository contains the weights of the TimeMoE-50M model of the paper Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts. For details on how to use this model, please visit our GitHub page.
News (2025.08) Sundial has been integrated into Apache IoTDB, a native time-series database. News (2025.06) Sundial has been accepted as ICML 2025 Oral (Top 1%). News (2025.05) Get 1st MASE on the GIFT-Eval Benchmark. News (2025.02) Get 1st MSE/MAE zero-shot performance on Time-Series-Library datasets. Sundial is a family of generative time series foundation models. This version is pre-trained on 1 trillion time points with 128M parameters. For more information, please refer to this paper. [[Slides]](https://cloud.tsinghua.edu.cn/f/8d526337afde465e87c9/) [[Poster]](https://cloud.tsinghua.edu.cn/f/cc2a156315e9453f99b3/) [[Intro (CN)]](https://mp.weixin.qq.com/s/y3sc2e2lmW1sqfnoK-ZdDA). The…