01 Home02 Research03 Projects04 Notes05 About06 Lab
Projects/Kronos evaluation

A forecast model,
tested against the simple.

An independent evaluation of the public Kronos-small checkpoint and an exact speed-up of its inference path. Kronos itself is an upstream model by its original authors.

My role
Evaluation, tracing, KV cache
Upstream
Kronos model and checkpoints by its original authors
Period
Jul – Sep 2026
Checkpoint
Kronos-small (public)
eval/topk_backtest.py · CSI300 · Kronos-small
Cumulative excess return over CSI300 after costs: Kronos-small ends at plus 1.3 percent, reversal plus 0.7, random minus 2.6, momentum minus 16.3.
Bounded KV cache11.7× fasterBit-identical output · CPU, 400-bar lookback
Ranking signalp = 0.295Not significant (Newey–West, 5-path run)

Redrawn from the evaluation’s daily reports: compounded excess over CSI300 after costs. The annualized mean, +1.93%, is the figure quoted elsewhere.

Question

Does the public checkpoint add value over simple baselines?

I separated forecast behaviour from ranking and portfolio performance, then compared Kronos-small with random, momentum and reversal strategies under one backtest setup.

My role

Evaluation, tracing and an exact speed-up.

I traced the tokenizer and autoregressive prediction path, built a real-data walkthrough, ran the evaluation and implemented a bounded KV cache: 11.7 times faster on CPU (400-bar lookback, 100 greedy steps) with bit-identical output, for contexts within the model’s valid length. I did not pretrain or author Kronos.

Evidence

A modest result, kept in context.

With 10 averaged forecast paths, Kronos-small recorded 1.93% annualized excess return after costs; the reversal baseline recorded 1.66%. A five-path run of the same setup gave Kronos 6.01%, and its ranking signal was not statistically significant (Newey–West p = 0.295). The chart keeps the weaker random and momentum comparisons.

CSI300 annualized excess return after modeled costs: Kronos-small plus 1.93 percent, reversal plus 1.66 percent, random minus 2.79 percent, momentum minus 18.80 percent.
Public Kronos-small checkpoint, CSI300, 225 dates from July 2024 to June 2025, 10-day forecast horizon, 10 averaged forecast paths and 0.15% cost on each buy and sell. Download the plotted metrics ↧
Limits

One public-checkpoint study.

  • One checkpoint, one universe, one window, one cost model.
  • Not a numerical reproduction of the paper, whose best model, Kronos-large, is not public.
  • No claim of general superiority or live performance.