English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Kronos: The First Open-Source Foundation Model for Financial Markets — Treating K-Line as a Language

Forum topic · ✨步子哥 · 2026-08-04

Summary

Kronos is the first open-source foundation model purpose-built for financial markets, accepted at AAAI 2026. It tackles long-standing challenges in financial time-series forecasting—non-stationarity, high noise, and heavy-tailed distributions—by reframing OHLCV candlestick data as a discrete language. The model uses a two-stage framework: a specialized tokenizer that performs hierarchical (multi-scale) quantization of continuous OHLCV values into discrete tokens, followed by an autoregressive Transformer pretrained on these tokens, mirroring how GPT handles natural language. Pretrained on candlestick data from 45+ global exchanges, Kronos ships in four sizes (mini at 4.1M parameters with 2k context, plus small, base, and a closed large variant). The paper (arXiv:2508.02739) and code are publicly available. Kronos illustrates a broader paradigm—"treating X as a language"—in which any continuous signal that can be tokenized becomes amenable to Transformer pretraining, extending beyond text to vision, audio, proteins, and now financial series.

Why Financial Time-Series Prediction Is Hard

Financial time-series forecasting has long resisted general-purpose models. Traditional approaches (ARIMA, LSTM) treat price as a continuous scalar, but markets exhibit three stubborn properties:

  • High noise — most price movement is noise rather than signal.
  • Non-stationarity — distributions drift over time; yesterday's edge vanishes tomorrow.
  • Heavy tails — extreme events occur far more often than a Gaussian assumption predicts.
  • General time-series foundation models (TSFMs) trained on cleaner domains (electricity load, traffic flow) tend to underperform on markets because they never learn the financial "dialect."

    Kronos: A Two-Stage Framework

    Kronos (GitHub Trending +217 stars/day, accepted at AAAI 2026, arXiv: 2508.02739) is the first open-source foundation model pretrained exclusively for financial markets. Its central innovation is a two-stage pipeline.

    Stage 1 — Hierarchical Tokenization of OHLCV

    OHLCV (open, high, low, close, volume) is continuous and multi-variate. Rather than uniform binning—which collapses a 1% move and a 5% move into the same bucket and loses multi-scale structure—Kronos applies hierarchical quantization: coarse tokens capture trends, fine tokens capture intra-bar volatility. The tokenizer encodes multiple temporal scales simultaneously, mirroring how human technicians read daily, weekly, and monthly charts side by side.

    Stage 2 — Autoregressive Transformer Pretraining

    On top of the hierarchical discrete tokens, a standard autoregressive Transformer is pretrained, GPT-style. The architecture is off the shelf; the innovation is the new *vocabulary*—the way continuous financial data is translated into tokens.

    Data and Model Family

    Pretraining data spans candlesticks from 45+ global exchanges, exposing the model to diverse market microstructures rather than a single national "dialect."

    | Model | Parameters | Context Length | Open Source | |---|---|---|---| | Kronos-mini | 4.1M | 2048 | ✅ | | Kronos-small | 24.7M | 512 | ✅ | | Kronos-base | 102.3M | 512 | ✅ | | Kronos-large | 499.2M | 512 | ❌ |

    The mini variant is striking: 4.1M parameters with 2k context is tiny by GPT norms, suggesting the "language" of financial series has a smaller vocabulary, simpler grammar, and shorter dependencies than natural language.

    The Deeper Trend — "X as a Language"

    Kronos exemplifies a widening pattern in modern AI: any signal that can be discretized into tokens becomes a candidate for Transformer pretraining.

  • NLP GPTs treat text tokens as language.
  • VLMs discretize image patches into tokens.
  • Speech models discretize waveforms into tokens.
  • Kronos discretizes candlesticks into tokens.
  • The common thread is discretize, then pretrain. Transformer's real strength is not natural language per se—it is sequence modeling over any well-designed token space.

    Cross-Paper Convergence

    Kronos echoes several contemporary threads:

  • Native multimodal scaling laws — different modalities need their own tokenizers; Kronos's financial tokenizer reinforces this.
  • Möbius RoPE and positional encoding — position encoding matters for temporal data; Kronos's context-length choices (2048 for mini, 512 for others) reflect explicit assumptions about temporal dependency depth.
  • Editing the construction plan, not the blueprint — innovation sits in the data representation layer, not the model architecture.
  • Resources

  • Project: https://github.com/shiyu-coder/Kronos
  • Paper: https://arxiv.org/abs/2508.02739
  • Live Demo: https://shiyu-coder.github.io/Kronos-demo/
  • Models on HuggingFace: NeoQuasar/Kronos-{mini,small,base}
  • License: open source
  • Venue: AAAI 2026

Tags

#kronos#foundation-model#financial-time-series#transformer#tokenization#ohlcv#aaai-2026#open-source

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178503928