- Run record
- loading
of Run 1, Boardly — Website Relaunch, Version 1.0.0
The first Revision the Run produced from the frozen inputs. It was produced by Codex CLI codex-cli 0.155.1 running GPT-6-Luna, in Run 1 of Challenge Version 1.0.0.
Published sites are independent, AI-generated demos served from separate preview origins with local mock data. They are not products, and no UIBenchmark session data crosses to them.
Which frozen inputs this Revision answers, and where it sits in its Run.
The immutable snapshot of the coding workflow. A changed tool version or exposed setting becomes a new Configuration rather than an edit of this one.
The model family page lists every Configuration recorded under this display name. It groups them for navigation only and never merges two identities.
What the operator recorded while the Run that produced this Revision was executed.
Attempts that failed for a technical reason and were discarded and retried from the same frozen inputs. The count is not a quality signal.
Generation cost: Not separately metered. Runs were executed on the operator’s personal tool subscriptions, which do not itemize a per-Run charge. No amount is attributed, and no zero-cost claim is made.
Three different moments, kept apart on purpose. All values are Unix milliseconds UTC in the archive and are shown in UTC here.
How long the Run took, as the operator recorded it. It is not a start or end moment.
When the uploaded Bundle became immutable in the archive.
The archive does not record a separate publication moment yet, so this stays unrecorded rather than reusing the Finalize time.
Token counts as the coding tool itself reported them. Nothing here is a cost, a price, or a measurement the platform took.
| Category as the tool reports it | Tokens |
|---|
What the automated checks recorded, and which evidence artifacts the operator approved for publication.
| Name | Kind | Size |
|---|---|---|
The recorded digests of the immutable trees. They are shown in full so a published Revision can be checked against the bytes it was built from.
Revision 0 is the only published Revision of Run 1.
Where the generated site is served from, and whether it is being served right now.
Each generated site is served from its own preview address under this site's domain. Sign-in cookies for this site are locked to its own address and are not sent to preview sites, but because previews share the domain a small cookie risk remains.
A Deployment is served only when the operator asked for it and the preview Worker confirmed the activation record.
Operator-attested: the Configuration, tool version, displayed model, and elapsed time were recorded by the operator from what the tool showed at Run time. They are not independently verified. An exact model ID appears only when the tool exposed one; it is never inferred.
Source is browser-viewable only. There is no download or archive endpoint, and no additional license is granted beyond inspection and research.
Index 0 is always the initial Revision and has no parent.
Whether the archive publishes this record, kept apart from whether its site is served right now.
| input_tokens | 395,884 |
|---|
| cached_input_tokens | 334,336 |
|---|
| cache_write_input_tokens | 0 |
|---|
| output_tokens | 37,279 |
|---|
| reasoning_output_tokens | 13,458 |
|---|
| total_tokens | 433,163 |
|---|
Tool-self-reported: token counts come from the coding tool’s own output, imported from a labeled local source. Category names are kept as the tool reports them and are not normalized between tools. Missing usage stays unknown. Nothing here claims equal compute or equal token budgets.
| build.log |
| evidence |
| 875 bytes |
| evaluation.json | evaluation report | 464.6 KiB |
| final-response.txt | evidence | 572 bytes |
| prompt.txt | evidence | 4.6 KiB |
Automated checks are not an accessibility certification. Recorded axe findings describe what the tooling observed at capture time, on a pinned browser and viewport. A check that reports no violation is not proof that the interface is accessible, and a selector problem is inconclusive rather than an implementation failure.