Cloud baseline

DeepSeek Pro Latest — 0.6591 on Gauntlet 7.2

~deepseek/deepseek-pro-latest · scorer sb-7.2

DeepSeek Pro Latest, single model via OpenRouter~deepseek/deepseek-pro-latest · single agentOct 4, 2026
Overall
0.6591
Excellent
no
Benchmark
Gauntlet 7.2
Model build
62m 48s
Prompt tok
29.5M
Gen tok
270.1k

Recorded inference cost

$0.63

From the saved OpenRouter generation billing summary (142 recorded requests). Includes auxiliary model calls made during the build. Scoring reuses the saved build; this amount describes the original inference spend.

  • Billing recovered in a subsequent one-time historical lookup on 2026-10-18. All original generation records were found; the original saved summary remains archived.

Token usage

Tokens: 29.5M in · 270k out · 26.4M cached (exact, OpenRouter records)

29,748,194 tokens over 142 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: 176,738 (not added to the total).

  • deepseek/deepseek-v4.1-flash-20260910

    In 19.5M · Out 90k · Cached 19.2M · Reasoning 56.6k · 65 calls

  • deepseek/deepseek-v4-pro-20260813

    In 10M · Out 180k · Cached 7.2M · Reasoning 120k · 76 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 contract0.88
D · finesse0.90

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 × 0.8387 core + 0.12 × 0.84 gate × 0.65 excellence) × 1.0000 critical = 0.8039

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 (13 of 16 met here), then pays out at the excellence tier's own measured mean. Core 0.7381 + excellence 0.0658.

j_first_use 1.00j_workflow_journey 1.00j_error_state 0.00j_empty_state 1.00console_clean 2.00v_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 0.50p_under_stream 1.00p_api_latency 1.00p_sync_wall 1.00

Earned score 0.804 · Admission ceiling 0.669 · Final score 0.659

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

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

Tower structure, currency mapping and scene truth: s_tower_geometry, s_currency_collar (maximum 0.669)

3D interaction and overview legibility: t_coast_identity, t_coast_reality, q_overview_legibility, q_inspector_framing (maximum 0.709)

Event animation and backend recovery: m_committed_event_replay (maximum 0.899)

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
deepseek14129,477,104270,11353680.181.3
gemini197251138.240.0
fleet14229,478,076270,11853451.181.1

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.

1First captured render
2Latest captured render
3After sync
4Error state
5Mobile · 375px

Run details

Model
~deepseek/deepseek-pro-latest
Engine events
0
Repair rounds
0
Started
Oct 2, 2026, 11:44 PM
Finished
Oct 3, 2026, 12:57 AM
Re-scored
Oct 4, 2026, 05:11 AM (same build, corrected scorer)

How the benchmark is scored