Cloud baseline

Opus 5.5 — 0.7926 on Gauntlet 7.2

anthropic/claude-opus-5.5 · scorer sb-7.2

Claude Opus 5.5, single model via OpenRouteranthropic/claude-opus-5.5 · single agentOct 4, 2026
Overall
0.7926
Excellent
no
Benchmark
Gauntlet 7.2
Model build
56m 48s
Prompt tok
5.8M
Gen tok
272.6k

Recorded inference cost

At least $7.34

From the saved OpenRouter generation billing summary (63 recorded requests). This is a lower bound; the billing record is incomplete. Includes auxiliary model calls made during the build. Scoring reuses the saved build; this amount describes the original inference spend.

  • The harness counted tokens no recovered generation id accounts for: prompt -972, completion -5

Token usage

Tokens: 5.8M in · 273k out · 5.7M cached (exact, OpenRouter records)

6,085,668 tokens over 63 model calls: every call of the run, the model's own and the harness's helper calls, counted from OpenRouter records. Reasoning tokens as the provider reports them: 185,688 (not added to the total).

  • anthropic/claude-opus-5.5-20260921

    In 5.8M · Out 273k · Cached 5.7M · Reasoning 186k · 62 calls

  • google/gemini-2.5-flash

    In 972 · Out 5 · Cached 0 · Reasoning 0 · 1 call

Tier breakdown

A · structure & runtime1.00
B · behaviour1.00
C · vendor contract1.00
D · finesse0.93

Scoring detail

How this score was built

Per-check scorer output, as posted by the app. Expand a section to inspect its checks.

Earned credit before admission

(0.88 × 0.9864 core + 0.12 × 0.98 gate × 1.00 excellence) × 0.8857 critical = 0.8732

Critical defects compound a multiplier on the whole score (pre-severity 0.9859):

j_workflow_journey 0.71 → ×0.89

The core (88% of the total) is the weighted mean of the thirteen measured tiers. The last 12% is the excellence slice: it unlocks in proportion to the perfection conditions below (15 of 16 met here), then pays out at the excellence tier's own measured mean. Core 0.8680 + excellence 0.1179.

j_first_use 1.00j_workflow_journey 0.71j_error_state 1.00j_empty_state 1.00console_clean 0.00v_responsive_375 1.00v_dates_readable 1.00t_scene_binding 1.00x_conservation_residual 1.00r_no_row_loss 1.00p_drag_frames 1.00p_idle_flatness 1.00p_stream_apply 1.00p_under_stream 1.00p_api_latency 1.00p_sync_wall 1.00

Earned score 0.873 · Admission ceiling 0.799 · Final score 0.793

Final score is the lower of earned credit and the admission ceiling. Passing admission adds no points.

Visible: passedMatching: passedGood: not metVisual excellence: not metBackend excellence: passed

3D interaction and overview legibility: t_labels_culling (maximum 0.799)

Event animation and backend recovery: an earlier admission band is incomplete (maximum 0.899)

Run notes and corrections

Scoring note, 2 November 2026. This run was graded by the sb-7.2 scorer as it was before harness version 3.0.148. That scorer only read notification text from elements with nothing nested inside them, unless the feed used a class name containing "notif". This build puts the kind in its own element inside each entry, so the scorer read none of its notifications and couldn't tell whether one appeared after the approve and reject steps. The notification parts of j_workflow_journey and j_workflow_reject may therefore understate how the model did. This build's files weren't kept, so it can't be re-graded with the fixed scorer, and the score stays as published.

Token rates

Measured by the engine itself, one record per completed model call: prefill rate is prompt tokens over time-to-first-token, generation rate is completion tokens over the decode window. Medians per node.

NodeCallsPrompt tokGen tokPrefill tok/sGen tok/s
claude625,812,099272,59214683.491.3
gemini197251394.589.3
fleet635,813,071272,59714416.691.1

Graded browser recording

Loading duration…

Watch the full graded browser recording at its original speed: the payment field, structure inspection, committed updates and replay. The check results and screenshots on this page provide the wider evidence. Pause or scrub to inspect a view.

Each tower represents a payment: height encodes its amount, cap color its status, and the collar its currency. The selected detailed tower is expected to animate its collar after a committed backend update or replay. These are the task requirements; the check results show what this app actually achieved.

Ready to play0:00 / —:—
Open recording file

Drag the timeline to seek. Keyboard: arrow keys move 5 seconds; Home and End jump to the start and end.

Full graded browser recording: payment field, structure inspection, committed updates and replay

Screenshots

Captured from the built application during browser grading. Captions identify the recorded views and checks; these images do not represent repair rounds.

1Initial app view
2Error state
3Payment workflow
4First captured render
5Latest captured render

Run details

Model
anthropic/claude-opus-5.5
Engine events
0
Repair rounds
0
Started
Oct 4, 2026, 10:37 AM
Finished
Oct 4, 2026, 11:40 AM

How the benchmark is scored