hw-apex-bench v0.2.0 · Evidence, not estimates

refrel3 / q16k / access

Synthetic evidence establishes correctness, not performance. Unknown machines remain separate; missing values are not zeros. Axis 5 v1 supports synthetic experiments only.

Leaderboard row

{
  "D_Q_Bps": 2084577899.5246098,
  "Q_bytes": 16384.0,
  "axis": 3,
  "evidence": "evidence/991080498c4f35d061a889bfedb7d6650293d6d0855b19bed3817580b53f294a/evidence.json",
  "evidence_sha256": "991080498c4f35d061a889bfedb7d6650293d6d0855b19bed3817580b53f294a",
  "format": "refrel3",
  "kind": "measured",
  "machine": "ace-core",
  "p50_us": 14.648,
  "p95_us": 27.551,
  "p99_us": 41.117,
  "scope": "cpu-in-process",
  "status": "CALIBRATION_ONLY",
  "table": "axis3-W1",
  "variant": "q16k",
  "verified": 10000,
  "window_bytes": 1
}

Exact evidence JSON · SHA-256 991080498c4f35d061a889bfedb7d6650293d6d0855b19bed3817580b53f294a

All evidence fields

Every evidence field is in the exact evidence JSON (SHA-256 991080498c4f35d061a889bfedb7d6650293d6d0855b19bed3817580b53f294a, 63998939 B), stored once and shared by every row of that evidence.

Hash-bound files

Shared input files retain their manifest associations; association is not a machine inference.

External data-root files (referenced by hash, not copied)

Reproduce and independently verify

Run in the applied benchmark clone, with this complete site at ./site. The output directory must be new. These commands verify existing evidence and regenerate HTML; they do not rerun measurements. DATA_ROOT is the release-input root holding the referenced catalog directories with exactly the listed bytes.

set -euo pipefail
python3 -m tools.release_v020 --verify site/downloads/release --manifest-sha256 8208dc3d1f7f07d0b8a345db239dd8076ff151f0ed93c8bfe0ca6b90d517d774 --data-root DATA_ROOT
python3 -m tools.build_site site/downloads/release --out rebuilt-site --manifest-sha256 8208dc3d1f7f07d0b8a345db239dd8076ff151f0ed93c8bfe0ca6b90d517d774 --data-root DATA_ROOT