reconbench

Indexsquoosh-image-compressionrun

codex/gpt-5.6-terra

20260830T100813Z-85bb41

completedtrusted

Verdict

Overall84.9
Score arithmetic
ComponentValueWeightContribution
Submission + build gatesubmission contract + install + build; policy is gated below✓ passedgate—
Global visual0.8308× 0.50+0.4154
Weighted regions1.0000× 0.00+0.0000
Functional0.3333× 0.10+0.0333
Responsive1.0000× 0.10+0.1000
Workflow1.0000× 0.30+0.3000
Preservation1.0000× 0.00+0.0000
Policy1.0000gate—
Overall= 0.8487

Visual evidence

Fig. 01desktopbase target · 1440×900

visual 77.1
reference capture for desktop
reference
candidate capture for desktop
candidate

pixel 93.9thumbnail 95.9fg-IoU 96.1edge 21.8dimension 100.0

Fig. 02mobilebase target · 390×844

visual 92.4
reference capture for mobile
reference
candidate capture for mobile
candidate

pixel 92.1thumbnail 93.8fg-IoU 97.2edge 84.9dimension 100.0

Fig. 03desktopbase target · 1440×900

visual 95.8
reference capture for desktop
reference
candidate capture for desktop
candidate

pixel 98.0thumbnail 98.4fg-IoU 100.0edge 86.1dimension 100.0

Fig. 04desktopbase target · 1440×900

visual 95.9
reference capture for desktop
reference
candidate capture for desktop
candidate

pixel 97.9thumbnail 98.3fg-IoU 100.0edge 86.7dimension 100.0

Fig. 05mobilebase target · 390×844

visual 81.9
reference capture for mobile
reference
candidate capture for mobile
candidate

pixel 96.7thumbnail 97.6fg-IoU 63.3edge 83.5dimension 100.0

Workflow replay

6 / 6 passed · weight 3.0 / 3.0

ResultCheckKindDimensionWeightDetail
image-compression-desktop.upload-source—workflow1.0—
image-compression-desktop.loaded-editor—workflow0.1—
image-compression-desktop.quality-50—workflow0.8—
image-compression-desktop.quality-result—workflow0.1—
image-compression-mobile.upload-source-mobile—workflow0.9—
image-compression-mobile.mobile-loaded—workflow0.1—

Deterministic checks

4 / 6 passed · weight 2.0 / 3.0

ResultCheckKindDimensionWeightDetail
required-file:submission.json—gate0.0—
source-upload-desktopselectorfunctional0.5Locator.wait_for: Timeout 5000ms exceeded. Call log: - waiting for locator("[aria-label='Image upload']").first to be visible
source-upload-mobileselectorfunctional0.5Locator.wait_for: Timeout 5000ms exceeded. Call log: - waiting for locator("[aria-label='Image upload']").first to be visible
squoosh-homeselectorfunctional0.5—
mobile-no-overflowno_horizontal_overflowresponsive1.0—
mobile-samplesselectorresponsive0.5—
Apparatus · agent, timing, usage and digests

Agent

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

Variant & network

Variant
—
Network policy
deny
Enforced
true

Timing & usage

Started
2026-08-30 10:08:14 UTC
Finished
2026-08-30 10:18:03 UTC
Elapsed
9m 49s
Input tokens
3.03M
Output tokens
23.4k
Cost
—

Content digests

evaluator_environment
fb0d05b65ef3dfe67d61dcd60b2a259135f2dae2641843e5543c1fa04be46e9c
grader
6c4737b6bfdf1979112b778e497c56783dfe12bf85502aed630d8d7c545871b2
oracle
31eaf13c43733562d1d44a6ae41d65d7adb925e41ebd23416e5632e6eb7701dc
runtime_config
9d5c6935b9046cdb4ecaafd1c77f8de58d49052dccdd3847bc93628ea2aee04e
submission
b77d43eb5e26d997a87cfe39b156ef46c8cab4337b5f103f87552e59d268a93c
task
eaa2e0896d8ea8b86948a5bcf9e258df064ab9bab43c2d2631455329298d24f0

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.