Cloud baseline

Sonnet 5 — 0.3751 on Gauntlet 7.0

us.anthropic.claude-sonnet-5 · scorer sb-7.0

Claude Sonnet 5Aug 22, 2026
Overall
0.3751
Excellent
no
Benchmark
Gauntlet 7.0
Recorded duration
33m 12s

Token usage

Tokens: at least 4.4M in · 149k out (lower bound, harness session counters)

4,568,174 tokens over 57 model calls: every call of the run, the model's own and the harness's helper calls, counted from harness session counters. These are a lower bound.

The harness's session counters for this run (the usage AWS Bedrock returned per response, summed); calls counted from the stored transcript (approximate). Prompt caching was off on Bedrock for Claude then, so no cached tokens. Bedrock keeps no per-call bill to query.

  • us.anthropic.claude-sonnet-5

    In 4.4M · Out 149k · 57 calls

Tier breakdown

A · structure & runtime0.91
B · behaviour0.96
C · vendor contract0.71
D · finesse0.70

Scoring detail

How this score was built

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

The number, exactly

(0.88 × 0.6892 core + 0.12 × 0.60 gate × 0.26 excellence) × 0.6000 critical = 0.3751

Critical defects compound a multiplier on the whole score (pre-severity 0.6252):

j_workflow_journey 0.00 → ×0.60

The core (88% of the total) is the weighted mean of the ten measured tiers. The last 12% is the excellence slice: it unlocks in proportion to the perfection conditions below (9 of 16 met here), then pays out at the excellence tier's own measured mean. Core 0.6065 + excellence 0.0187.

j_first_use 0.60j_workflow_journey 0.00j_error_state 0.00j_empty_state 1.00console_clean 12.00v_responsive_375 1.00v_dates_readable 1.00t_scene_binding 0.00x_conservation_residual 1.00r_no_row_loss 1.00p_drag_frames 0.00p_idle_flatness 1.00p_stream_apply 0.00p_under_stream 1.00p_api_latency 1.00p_sync_wall 1.00

Run notes and corrections

Cloud baseline — a single harness run session on the frozen sb-7 spec (Meridian Payments Console: two services, 12,288-payment collection, signed webhooks, maker/checker approval workflow, instanced WebGL 3D field), scored by the sb-7.0 scorer. Screenshots are the scorer's own render-gate captures of the built app. Full sb-7 tier means (the schema's tierA–tierD carry only A–D): A 0.9083 · B 0.9590 · C 0.7143 · D 0.7000 · J 0.5143 · V 1.0000 · P 0.6667 · T 0.0774 · X 0.8333 · R 0.8817 · E 0.1560.

Screenshots

Recorded views of the built application. Captions identify the available captures; the number of images does not imply a number of repair rounds.

1First render — before repairs
2Final render
33D field · 12,288 instanced columns · WebGL
4Approval workflow · drafts panel
5After sync

Run details

Model
us.anthropic.claude-sonnet-5

How the benchmark is scored