reconbench

Indexstackedit-field-notesrun

codex/gpt-5.6-luna

20260830T114500Z-38c2b6

completedtrusted

Verdict

Overall81.9
Score arithmetic
ComponentValueWeightContribution
Submission + build gatesubmission contract + install + build; policy is gated below✓ passedgate—
Global visual0.6375× 0.50+0.3188
Weighted regions1.0000× 0.00+0.0000
Functional1.0000× 0.10+0.1000
Responsive1.0000× 0.10+0.1000
Workflow1.0000× 0.30+0.3000
Preservation1.0000× 0.00+0.0000
Policy1.0000gate—
Overall= 0.8188

Visual evidence

Fig. 01desktopbase target · 1440×900

visual 87.3
reference capture for desktop
reference
candidate capture for desktop
candidate

pixel 98.0thumbnail 98.9fg-IoU 75.2edge 86.5dimension 100.0

Fig. 02mobilebase target · 390×844

visual 82.3
reference capture for mobile
reference
candidate capture for mobile
candidate

pixel 96.6thumbnail 97.2fg-IoU 65.2edge 82.9dimension 100.0

Fig. 03mobile-readerstate on mobile

visual 88.6
reference capture for mobile-reader
reference
candidate capture for mobile-reader
candidate

pixel 96.7thumbnail 97.4fg-IoU 83.4edge 82.2dimension 100.0

Fig. 04desktopbase target · 1440×900

visual 87.2
reference capture for desktop
reference
candidate capture for desktop
candidate

pixel 98.9thumbnail 99.2fg-IoU 76.9edge 82.4dimension 100.0

Workflow replay

5 / 5 passed · weight 2.0 / 2.0

ResultCheckKindDimensionWeightDetail
edit-persistence.replace-note—workflow0.6—
edit-persistence.replace-title—workflow0.4—
edit-persistence.reload—workflow0.8—
edit-persistence.restore-title—workflow0.1—
edit-persistence.persisted—workflow0.1—

Deterministic checks

7 / 7 passed · weight 4.5 / 4.5

ResultCheckKindDimensionWeightDetail
required-file:submission.json—gate0.0—
seeded-previewtextfunctional1.0—
editor-presentselectorfunctional0.5—
reader-controlselectorfunctional0.5—
document-titleselectorfunctional0.5—
mobile-no-overflowno_horizontal_overflowresponsive1.0—
mobile-preview-hiddenhiddenresponsive1.0—
Apparatus · agent, timing, usage and digests

Agent

Selector
codex/gpt-5.6-luna
Runtime
codex · Docker version 29.2.1, build a5c7197
Model
gpt-5.6-luna
Effort
high
Tool profile
baseline

Variant & network

Variant
—
Network policy
deny
Enforced
true

Timing & usage

Started
2026-08-30 11:45:01 UTC
Finished
2026-08-30 11:56:11 UTC
Elapsed
11m 09s
Input tokens
1.93M
Output tokens
31.2k
Cost
—

Content digests

evaluator_environment
fb0d05b65ef3dfe67d61dcd60b2a259135f2dae2641843e5543c1fa04be46e9c
grader
6c4737b6bfdf1979112b778e497c56783dfe12bf85502aed630d8d7c545871b2
oracle
b65c5852c9f91526271059dff43b1bcbdc3096516912e18bd93152a0cdafd081
runtime_config
9be2f291cf148821361bf9e22eb4aa3f6d242f4063f2860b53f9031f62c8baed
submission
34b9c591a0d8522ea7f7b5659bb053f29ebbad405c6414ba3a4a7ac20ab696c4
task
5a9a10ac23a51761eb1c771a22309ac235f4f86aeb670a7a131198606f30c223

Digests pin the exact task package, submission tree, oracle, grader and evaluator environment used. They are how a result is re-verified without trusting this page.