Measured harness ledgerPublic result
Grok 4.6Sekiro Main Menu — Grok 4.6 xhigh
Recreate a high-end cinematic Sekiro-style main menu in the browser from the pinned reference and task contract.
xhigh reasoningHeadline result
- Workflow cost
- $1.15
- Wall-clock
- 51m 59.543s wall-clock
- Processed tokens
- 9.42M processed
- Record state
- partial_two_turn_metrics_source_build_and_static_serve_ledger
Public summary
Grok 4.6 xhigh partial_two_turn_metrics_source_build_and_static_serve_ledger ledger: 51m 59.543s wall-clock, 9.42M processed, and $1.15 Provider-recorded usage estimate, not a subscription cash-charge invoice.
Run identity and stack
- Result ID: sekiro-main-menu-browser-grok-4.6-xhigh
- Technical model: Grok 4.6
- Provider: xAI Grok Build
- Client: Grok 1.0.5
- Stack: xAI Grok Build
- Stack: Grok 1.0.5
- Stack: Technical model/configuration: Grok 4.6
- Stack: Browser recreation
- Stack: Pinned visual reference
- Stack: Harness v1 Sekiro prompt
- Stack: Requested tool profile: Grok Build CLI
Cost basis
- The exact billed alias grok-4.6-build has no official public catalog price, so no API-equivalent cost is inferred.
Primary artifact integrity
- Kind: manifest-verified-artifact
- Path: artifacts/sekiro-main-menu-browser-grok-4.6-xhigh/production-build/index.html
- SHA-256: 61d77b2e5a8836a79f9a530bd0bb4dc7b131a03a4f4dc1f4d48a4910e29b4a1c
Recorded caveats
- Wall-clock is end-to-end workflow latency, not model-only compute; it includes tools, installs, browser checks, and idle intervals.
- The benchmark protocol includes two genuine ACP user turns, including a host task-completed notification. It is a disclosed protocol deviation, not a one-shot result.
- Generated output includes visible output and reasoning output. Cache reads are discounted, so total processed tokens overstate effective cost.
- The $1.15 figure is the source ledger's provider-recorded usage estimate, not an inferred API-equivalent rate or a subscription cash charge.
- The fresh build and static HTTP checks validate basic runnability only; they are not browser interaction, keyboard navigation, audio, local-FPS, or final-render measurements.
- No RTX Pro 6000 final render, local FPS/environment record, supplied final capture, or blind evaluation is archived.
Visible evidence gaps
- local FPS measurement with browser, hardware, and viewport metadata
- keyboard navigation and menu interaction record
- RTX Pro 6000 final render or capture metadata
- blind-evaluation record
Public result only
This result keeps its public summary and evidence, but it does not currently have a matching Builder test with prompts, projects, Harness workflows or skills.