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…
Open-weight model · Time series forecasting
Kronos-Tokenizer-base
by ShiYu NeoQuasar/Kronos-Tokenizer-base
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.
Parameters4M
Context—
Weights15.8 MB
Licensemit
AccessOpen weights
Monthly Downloads2.5M
Runs On
What it takes to serve Kronos-Tokenizer-base (4M 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 | 0.0 GB | 0.0 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 Kronos-Tokenizer-base
The memory line reads 0.0 GB at 16-bit, 8-bit and 4-bit, and that is no error. This is the tokenizer stage of Kronos, a decoder-only family built for financial candlesticks: it quantizes OHLCV K-line data into hierarchical discrete tokens for the autoregressive Transformer in stage two. At 4M parameters and 15.85 MB it will not decide your hardware. The cheapest option we list, a single MI300X with 192 GB at $1.85 an hour on-demand, is sized for the second stage, not this piece.
MIT is as light as licenses get: commercial use, modification and redistribution are permitted as long as the copyright and permission notices travel with the files. The check is scope. This page carries no context length and no configuration block, so get the second-stage model and its limits before planning a deployment. Training data came from over 45 global exchanges; confirm yours are covered.
Model Card
By ShiYu, published under mit, revision 0e0117387f39.
Kronos: A Foundation Model for the Language of Financial Markets
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.
Introduction
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.
Identity and Version
- Repository
- NeoQuasar/Kronos-Tokenizer-base
- Publisher
- ShiYu
- Task
- Time series forecasting
- Modality
- Time series
- Library
- pytorch
- Parameters
- 4M parameters
- Languages
- Not stated by the source
- Revision
- 0e0117387f39004a9016484a186a908917e22426
- First published
- 2025-06-30
- Last updated
- 2025-09-09
Files and Weights
4 files, 15.9 MB in total. The weights are 1 file totalling 15.8 MB in safetensors.
Weights1 file · 15.8 MB
Configuration1 file · 301 B
Documentation1 file · 8.9 KB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 15.8 MB | 59d85f6af76a |
| config.json | Configuration | 301 B | — |
| README.md | Documentation | 8.9 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 15.8 MB
Download from ShiYu
Released by ShiYu through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:2508.02739
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 15.8 MB |
| 16-bit | 0.0 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 Kronos-Tokenizer-base
Questions About Kronos-Tokenizer-base
How much GPU memory does Kronos-Tokenizer-base need?
About 0 GB at 16-bit and 0 GB at 4-bit: the weights (4M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Kronos-Tokenizer-base 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 Kronos-Tokenizer-base commercially?
Yes. Kronos-Tokenizer-base is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
Similar Models
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…
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…
Building on top of TTM-R1 and TTM-R2, we introduce the next generation of TinyTimeMixer under the Granite time-series foundation model family — Granite-TTM-R3. This release incorporates several novel tiny-neural architectural innovations designed to push the limits of accuracy in high-speed forecasting, a critical requirement for real-world production deployments. Granite-TTM-R3 is a family of pretrained models supporting multiple real-world forecasting scenarios: - Zero-shot forecasting across unseen datasets - Few-shot adaptation effective with as few as ~1K samples - Full fine-tuning for domain-specific optimization - Multivariate time-series forecasting - Exogenous / control variable…
TinyTimeMixers (TTMs) are compact pre-trained models for Multivariate Time-Series Forecasting, open-sourced by IBM Research. With model sizes starting from 1M params, TTM introduces the notion of the first-ever “tiny” pre-trained models for Time-Series Forecasting. The paper describing TTM was accepted at NeurIPS 24. TTM outperforms other models demanding billions of parameters in several popular zero-shot and few-shot forecasting benchmarks. TTMs are lightweight forecasters, pre-trained on publicly available time series data with various augmentations. TTM provides state-of-the-art zero-shot forecasts and can easily be fine-tuned for multi-variate forecasts with just 5% of the training…
TinyTimeMixers (TTMs) are compact pre-trained models for Multivariate Time-Series Forecasting, open-sourced by IBM Research. With less than 1 Million parameters, TTM (accepted in NeurIPS 24) introduces the notion of the first-ever “tiny” pre-trained models for Time-Series Forecasting. TTM outperforms several popular benchmarks demanding billions of parameters in zero-shot and few-shot forecasting. TTMs are lightweight forecasters, pre-trained on publicly available time series data with various augmentations. TTM provides state-of-the-art zero-shot forecasts and can easily be fine-tuned for multi-variate forecasts with just 5% of the training data to be competitive. Refer to our paper for…