Benchmark run

Qwen3.6 27B Fable Fusion, 3-node local fleet — 0.6618 on Gauntlet 5.2

qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp · 3 nodes · scorer sb-5.2

mighty-crane-54f23-node local fleetAug 17, 2026
Overall
0.6618
Hard
1.00
Excellent
no
Benchmark
Gauntlet 5.2
Recorded duration
93m 31s
Nodes
3

Token usage

Tokens: at least 1.3M in · 135k out (lower bound, harness session counters)

1,442,684 tokens over 240 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.

Local 3-node LM Studio fleet, no bill. Usage each node's server reported per call, summed from the harness's sessions for this run (188 sessions incl. 3 fix-attempt sessions); calls counted from the stored transcript (approximate). No cache counters were reported.

  • gabee-qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp

    In 370k · Out 39k · 57 calls

  • mihai-qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp

    In 500k · Out 40.9k · 85 calls

  • workhorse-qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp

    In 437k · Out 55.3k · 98 calls

Tier breakdown

A · structure & runtime1.00
B · behaviour0.38
C · vendor contract0.70
D · finesse0.81

Scoring detail

How this score was built

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

ComponentWeightEarnedPoints
Core (tiers A–D)60%0.6770.406
A · Structure25% of core1.000.150
B · Behaviour30% of core0.380.068
C · Vendor contract25% of core0.700.105
D · Finesse20% of core0.810.098
Journey (J)15%0.4500.068
Visual (V)10%0.5500.055
Performance (P)5%0.6670.033
Hard blocks10%1.0000.100
Total0.6618

Findings that held

Raised by the verify gate, still open when the run ended.

1

`pytest -q` failed — the generated tests exercise runtime paths that `--help`/`--collect-only` never invoke: elif self._mock_methods is not No ... [middle elided — head + tail shown] ... ary_after_sync - KeyError: 'created_at_utc' FAILED tests/test_api.py::test_sync_first_call - AssertionError: assert 0 == 60 FAILED tests/test_meridian.py::TestMeridianClient::test_create_payment_idempote

2

POST /api/sync is not CHEAP on a repeat run — the second sync re-fetched 247 row(s) it already had. FIX: make the client send If-None-Match per page — ... [middle elided — head + tail shown] ... ge's ETag keyed by (path, offset, limit) from the first sync and replay THAT page's ETag on the matching request; treat 304 as 'page unchanged, keep local rows'. One ETag replayed on every page never mat

Repair progression

R0·2R1·2

One chip per verify round: the number of open findings at that round. The run ends when a round reports zero findings or the round budget is exhausted.

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
3After sync
4Mobile · 375px

Run details

Model
qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp
Engine events
353
Repair rounds
1
Started
Aug 17, 2026, 06:30 AM
Finished
Aug 17, 2026, 08:04 AM

Fleet nodes

NodeModel
gabeegabee-qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp
mihaimihai-qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp
workhorseworkhorse-qwen3.6-27b-fable-fusion-711-uncensored-heretic-nm-dau-neo-max-mtp

How the benchmark is scored