Skip to content
Technology How it works Breeder — Hyperion Burner — Aegis Burner — MetroVolt AI-Native Architecture Magnets Fuel cycle Safety Roadmap
Solutions AI & Data Centers Defense & Government Grid & Baseload Neutron Detection Quantum
Learn Technical Library
Proof Publications Whitepapers Technical Library Open Science & Reproducibility The Honest Gates
Company About / Mission Leadership Environment Health & Safety Investors Careers Press Contact
3D Model
Technical LibraryValidation & Reproducibility
Validation & Reproducibility

Tier-1 and Tier-2 Reproduction

Not every correct calculation is bit-for-bit repeatable. Kronos states, per result, which standard applies.

Validation & ReproducibilityUpdated 2026-08-11

Deterministic, integer-and-float-stable calculations are graded Tier 1: on a matching software environment they reproduce byte-for-byte. Calculations that involve legitimate sources of numerical variation — iterative solvers, sampling, platform math differences — are graded Tier 2 and reproduce within a declared tolerance.

Grading each result this way is itself an honesty mechanism: it prevents a within-tolerance match from being over-sold as exact, and it tells a reviewer immediately whether a byte difference is a real discrepancy or expected noise.

Environment mattersTier-1 reproduction assumes the pinned software environment in the deposit; the reproduction notes specify it.

Common questions

What are Tier-1 and Tier-2 reproduction?

Tier-1 is byte-exact reproduction of deterministic results; Tier-2 is agreement within a stated tolerance where results are not bit-for-bit deterministic. Both are defined so 'reproduces' has a precise meaning.

Why distinguish them?

Because honesty about what 'reproduces' means matters — some computations match exactly, others within tolerance, and Kronos states which is which.

← PreviousThe Burner Validation SuiteNext →What Reproducibility Does and Doesn't Prove
Content reviewed August 2026 · design-and-simulation stage