# Kronos-mini by ShiYu: Open-Weight Model Source: https://savrn.com/models/kronos-mini Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve Kronos-mini (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](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.0 GB | 0.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.0 GB | 0.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 25, 2026. [Kronos-mini on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/kronos-mini/gpus) ## SAVRN's Notes on Kronos-mini Sixteen megabytes on disk, four files, 4M parameters: that is the entire footprint. Kronos-mini belongs to the Kronos family of decoder-only models pretrained on financial candlestick sequences from more than 45 exchanges, with a tokenizer that turns OHLCV data into discrete tokens. The memory table rounds to 0.0 GB at every precision, and the $1.85 per hour MI300X with 192 GB that heads the price list is a floor, not a requirement. Anything already in the rack will carry this. MIT permits commercial use, modification and redistribution with the notices kept, so building on it needs no license negotiation. The gaps in the record are what to check: no library or architecture is listed, no context length is recorded and no evaluation results are reported, so the paper at arXiv:2508.02739 is where the method is described. Weights ship as safetensors only, last updated September 9, 2025. SAVRN Research, 2026-09-18 ## Model Card By ShiYu, published under mit, revision f4e68697d9d5. ### 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. [Read the full model card (850 words)](https://savrn.com/models/kronos-mini/card) ## Identity and Version Repository NeoQuasar/Kronos-mini Publisher ShiYu Task Time series forecasting Modality Time series Library Not stated by the source Parameters 4M parameters Languages Not stated by the source Revision f4e68697d9d5aed55cef5c96aabc3376bcad9f81 First published 2025-07-01 Last updated 2025-09-09 ## Files and Weights 4 files, 16.5 MB in total. The weights are 1 file totalling 16.4 MB in safetensors. Weights1 file · 16.4 MB Configuration1 file · 225 B Documentation1 file · 8.8 KB Repository1 file · 1.5 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 16.4 MB | a7d5f37e2e9f | | config.json | Configuration | 225 B | — | | README.md | Documentation | 8.8 KB | — | | .gitattributes | Repository | 1.5 KB | — | ## License and Download License mit Access Open weights, no gate Download size 16.4 MB [Download from ShiYu](https://huggingface.co/NeoQuasar/Kronos-mini) Released by ShiYu through its official repository on Hugging Face. [Read the license](https://opensource.org/license/mit). ## Built From - Described by [arXiv:2508.02739](https://savrn.com/papers/kronos-a-foundation-model-for-the-language-of-financial-markets) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 16.4 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-mini - [Kronos-mini vs Kronos-Tokenizer-base](https://savrn.com/models/compare/kronos-mini-vs-kronos-tokenizer-base) ## Questions About Kronos-mini ### How much GPU memory does Kronos-mini 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-mini 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-mini commercially? Yes. Kronos-mini 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 Model · Time series forecasting ### [Toto-2.0-4m](https://savrn.com/models/toto-2-0-4m) [Datadog](https://savrn.com/model-publishers/datadog) 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… Open weights apache-2.0 4M parameters pytorch [View model](https://savrn.com/models/toto-2-0-4m) Model · Time series forecasting ### [Kronos-Tokenizer-2k](https://savrn.com/models/kronos-tokenizer-2k) [ShiYu](https://savrn.com/model-publishers/neoquasar) 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 weights mit 4M parameters torch [View model](https://savrn.com/models/kronos-tokenizer-2k) Model · Time series forecasting ### [Kronos-Tokenizer-base](https://savrn.com/models/kronos-tokenizer-base) [ShiYu](https://savrn.com/model-publishers/neoquasar) 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 weights mit 4M parameters pytorch [View model](https://savrn.com/models/kronos-tokenizer-base) Model · Time series forecasting ### [granite-timeseries-ttm-r3](https://savrn.com/models/granite-timeseries-ttm-r3) [IBM Granite](https://savrn.com/model-publishers/ibm-granite) 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… Open weights apache-2.0 1M parameters [View model](https://savrn.com/models/granite-timeseries-ttm-r3) Model · Time series forecasting ### [granite-timeseries-ttm-r2](https://savrn.com/models/granite-timeseries-ttm-r2) [IBM Granite](https://savrn.com/model-publishers/ibm-granite) 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… Open weights apache-2.0 805,280 parameters granite-tsfm [View model](https://savrn.com/models/granite-timeseries-ttm-r2) Model · Time series forecasting ### [granite-timeseries-ttm-r1](https://savrn.com/models/granite-timeseries-ttm-r1) [IBM Granite](https://savrn.com/model-publishers/ibm-granite) 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… Open weights apache-2.0 805,280 parameters granite-tsfm [View model](https://savrn.com/models/granite-timeseries-ttm-r1) ## ShiYu [All models and datasets](https://savrn.com/model-publishers/neoquasar) ## Versions - [f4e68697d9d5](https://savrn.com/models/kronos-mini/versions/f4e68697d9d5) · current 2026-09-18 ## Explore More - [All time series forecasting models](https://savrn.com/models/tasks/time-series-forecasting) - [All models under mit](https://savrn.com/models/licenses/mit) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-20. - [Hugging Face record](https://huggingface.co/NeoQuasar/Kronos-mini) - [How the hub is built](https://savrn.com/model-hub/methodology)