how-to-guide

How to Fine-Tune the Kronos Tokenizer for New Market Data: A Complete Guide

April 10, 2026 shiyu-coder/Kronos ↗

Fine-tune the Kronos tokenizer by preparing a CSV with six market features, configuring a YAML file with your data path and training epochs, and running finetune_csv/finetune_tokenizer.py to adapt the Binary Spherical Quantizer to your specific market distribution.

Adapting the Kronos tokenizer to new market data allows the model to learn domain-specific quantization patterns that improve downstream prediction accuracy. This process involves self-supervised fine-tuning of the Binary Spherical Quantizer (BSQuantizer) on your custom

Have a question about this repo?

These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:

Share the following with your agent to get started:

curl -s "https://instagit.com/install.md"

Add to your MCP client configuration:

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}

Ask your agent:

"Use Instagit MCP to understand how shiyu-coder/Kronos works."

Works with

Claude Codex Cursor VS Code OpenClaw Any MCP Client

Maintain an open-source project? Get it listed too →