Kronos is the first open-source foundation model for financial candlesticks (K-lines), trained on data from over 45 global exchanges. It is designed to handle the unique, high-noise characteristics of financial data. Kronos is a family of decoder-only foundation models, pre-trained specifically for the "language" of financial markets—K-line sequences. It leverages a novel two-stage framework: 1. A specialized tokenizer first quantizes continuous, multi-dimensional K-line data (OHLCV) into hierarchical discrete tokens. 2. A large, autoregressive Transformer is then pre-trained on these tokens, enabling it to serve as a unified model for diverse quantitative tasks. The success of large-scale…
This is the small variant of the Chronos-2 model with 28M parameters. For usage and details on the Chronos-2 model, please refer to https://huggingface.co/autogluon/chronos-2.
Parameters28M
Context—
Weights111.7 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads6.9M
Runs On
What it takes to serve chronos-2-small (28M 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 | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.0 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-small
There is one job here, forecasting a numeric series, and it makes the hardware question nearly disappear. With 28M parameters, Chronos-2-small needs 0.1 GB at 16-bit. The cheapest setup on this page is one MI300X with 192 GB at $1.85 an hour, so nobody should rent a card for this model alone. We would place it beside something larger and treat its memory as a rounding error.
Nothing in the license slows a deployment: Apache 2.0 permits commercial use, modification and redistribution, with an express patent grant, and asks that the notices travel with it and significant changes be stated. Two checks. No context length is listed, so settle the history window with the publisher's documentation. And the training relations name autogluon/chronos_datasets and Salesforce/GiftEvalPretrain; if your series overlap with either, your test numbers will flatter it. No evaluations are reported here, so run your own.
Model Card
By Autogluon, published under apache-2.0, revision ddec01313e50.
This is the small variant of the Chronos-2 model with 28M parameters. 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 the small variant of the Chronos-2 model with 28M parameters. 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-small
- Publisher
- Autogluon
- Task
- Time series forecasting
- Modality
- Time series
- Library
- chronos-forecasting
- Parameters
- 28M parameters
- Languages
- Not stated by the source
- Revision
- ddec01313e50b6bc58ebaa92ede81bc24a3d9f9a
- First published
- 2025-12-03
- Last updated
- 2025-12-03
Files and Weights
4 files, 111.8 MB in total. The weights are 1 file totalling 111.7 MB in safetensors.
Weights1 file · 111.7 MB
Configuration1 file · 969 B
Documentation1 file · 1.4 KB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 111.7 MB | 492290ae82bb |
| config.json | Configuration | 969 B | — |
| README.md | Documentation | 1.4 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 111.7 MB
Download from Autogluon
Released by Autogluon through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:2510.15821
- Trained on (disclosed) Salesforce/GiftEvalPretrain
- Trained on (disclosed) autogluon/chronos_datasets
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 111.7 MB |
| 16-bit | 0.1 GB |
| 8-bit | 0.0 GB |
| 4-bit | 0.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Compare chronos-2-small
Questions About chronos-2-small
How much GPU memory does chronos-2-small need?
About 0.1 GB at 16-bit and 0 GB at 4-bit: the weights (28M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run chronos-2-small 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-small commercially?
Yes. chronos-2-small 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
Toto (Time Series Optimized Transformer for Observability) is a family of time series foundation models for multivariate forecasting developed by Datadog. Toto 2.0 is the current generation, featuring u-μP-scaled transformers ranging from 4m to 2.5B parameters, all trained from a single recipe. Forecast quality improves reliably with parameter count across the family. The family sets a new state of the art on three forecasting benchmarks: BOOM, our observability benchmark; GIFT-Eval, the standard general-purpose benchmark; and the recent contamination-resistant TIME benchmark. Inference code is available on GitHub. For more examples, see the Quick Start notebook and GluonTS integration…
Update Feb 14, 2025: Chronos-Bolt models are now available on Amazon SageMaker JumpStart! Check out the tutorial notebook to learn how to deploy Chronos endpoints for production use in a few lines of code. Chronos-Bolt is a family of pretrained time series forecasting models which can be used for zero-shot forecasting. It is based on the T5 encoder-decoder architecture and has been trained on nearly 100 billion time series observations. It chunks the historical time series context into patches of multiple observations, which are then input into the encoder. The decoder then uses these representations to directly generate quantile forecasts across multiple future steps—a method known as…
Update Feb 14, 2025: Chronos-Bolt models are now available on Amazon SageMaker JumpStart! Check out the tutorial notebook to learn how to deploy Chronos endpoints for production use in a few lines of code. Chronos-Bolt is a family of pretrained time series forecasting models which can be used for zero-shot forecasting. It is based on the T5 encoder-decoder architecture and has been trained on nearly 100 billion time series observations. It chunks the historical time series context into patches of multiple observations, which are then input into the encoder. The decoder then uses these representations to directly generate quantile forecasts across multiple future steps—a method known as…
Update Feb 14, 2025: Chronos-Bolt & original Chronos models are now available on Amazon SageMaker JumpStart! Check out the tutorial notebook to learn how to deploy Chronos endpoints for production use in a few lines of code. Update Nov 27, 2024: We have released Chronos-Bolt models that are more accurate (5% lower error), up to 250 times faster and 20 times more memory-efficient than the original Chronos models of the same size. Check out the new models here. Chronos is a family of pretrained time series forecasting models based on language model architectures. A time series is transformed into a sequence of tokens via scaling and quantization, and a language model is trained on these…
Update Feb 14, 2025: Chronos-Bolt & original Chronos models are now available on Amazon SageMaker JumpStart! Check out the tutorial notebook to learn how to deploy Chronos endpoints for production use in a few lines of code. Update Nov 27, 2024: We have released Chronos-Bolt models that are more accurate (5% lower error), up to 250 times faster and 20 times more memory-efficient than the original Chronos models of the same size. Check out the new models here. Chronos is a family of pretrained time series forecasting models based on language model architectures. A time series is transformed into a sequence of tokens via scaling and quantization, and a language model is trained on these…