> ## Documentation Index
> Fetch the complete documentation index at: https://docs.inkwell.finance/llms.txt
> Use this file to discover all available pages before exploring further.

# Benchmarks

> Headline measured wall-clock for Dagon's match engine on a datacenter GPU, with the verification discipline behind every cited number.

> **CONFIDENTIAL & PROPRIETARY © 2026 Inkwell Finance, Inc. All Rights Reserved.**
>
> This document is for informational purposes only and does not constitute legal, tax, or investment advice, nor an offer to sell or a solicitation to buy any security or other financial instrument. Any examples, structures, or flows described here are design intent only and may change.

Every number on this page is tagged with a status — `measured`,
`measured-with-caveat`, `projected`, or `extrapolated-from-SOTA` — and
backed by a committed log. No figure on this page is an opaque
estimate. For the methodology that defines those tags, see
[Methodology](/dagon/performance/methodology).

## Headline

<CardGroup cols={2}>
  <Card title="N=32 match cycle · H100" icon="stopwatch">
    **7.77 s** wall-clock on an NVIDIA H100 SXM, with every output
    decrypted and asserted against the plaintext expected value.
    Correctness-preserving per-op programmable bootstrap.
  </Card>

  <Card title="Atomic homomorphic min · H100" icon="bolt">
    **15.1 ms** mean (== p50) over 20 iterations. **20/20**
    decryptions match the plaintext answer. The one-comparison floor
    — every higher-level op composes from it.
  </Card>
</CardGroup>

<Info>
  The headline numbers were measured on a fresh H100 SXM cloud pod.
  Inter-pod variance on this hardware is typically in the low-double
  digits of percent; the 7.77 s figure is a stable reference within
  that envelope.
</Info>

## Programmable-bootstrap headlines

The match cycle uses a programmable-bootstrap FHE backend — every
homomorphic primitive folds one bootstrap, so chain depth is not a
correctness constraint. Headline figures on the production target
(H100 SXM):

| workload               |       wall-clock | verification                        |
| ---------------------- | ---------------: | ----------------------------------- |
| Atomic `min(u32, u32)` | **15.1 ms** mean | 20/20 outputs verified              |
| N=32 match cycle       |       **7.77 s** | every output decrypted and asserted |

Verification runs on the host after `cudaStreamSynchronize`; its
wall-clock is reported separately from the match timer. Decrypt
overhead is well under one percent of the match wall-clock.

## CKKS reference (parity-fail exhibit)

A leveled CKKS path on the same hardware was run as a throughput
reference. At the bootstrap chain depth this matcher needs, CKKS does
not preserve correctness: the wall-clock measurement completes but
the decrypted answer diverges from the plaintext shadow beyond
tolerance. We cite this number only with the failure mode stated
inline; see
[Why CKKS fails at scale](/dagon/performance/why-ckks-fails) for the
structural reason.

## Next-generation FHE (projected — pre-alpha)

Production REFHE substrate is expected to deliver substantial
per-primitive speedup over current TFHE-style benchmarks once
mainnet alpha lands. Concrete margins will be measured at that
point. Any pre-mainnet figure for this backend is treated as
**`extrapolated-from-SOTA`** under the
[methodology discipline](/dagon/performance/methodology) — derived
from published primitive cost, not measurement.

## Verification scope

Every measured TFHE-style number on this page decrypts every output
after the match timer stops and asserts equality against the
plaintext expected value. A final invariant assertion in each
harness guards against silent undercounts. Decrypt wall-clock is
reported separately so reviewers can confirm verification runs
outside the match timer.

## Related

<CardGroup cols={2}>
  <Card title="Methodology" icon="list-check" href="/dagon/performance/methodology">
    The four status tags (measured / measured-caveat / projected /
    extrapolated) and the rules for citing each.
  </Card>

  <Card title="Why CKKS fails at scale" icon="triangle-exclamation" href="/dagon/performance/why-ckks-fails">
    What the CKKS H100 parity failure means, the absolute bootstrap
    noise floor, and why programmable-bootstrap is the
    correctness-preserving path.
  </Card>

  <Card title="SIMD amortization" icon="layer-group" href="/dagon/performance/simd-amortization">
    How packing takes single-batch latency down to per-batch
    amortized cost on a saturated GPU.
  </Card>

  <Card title="Interactive scaling dashboard ↗" icon="chart-line" href="https://dagon.so/scaling">
    Walk through the headline measured anchors interactively.
  </Card>
</CardGroup>