← Back to leaderboard  ·  Show all task instances

tomarrell/wrapcheck

A Go linter to check that errors from external packages are wrapped

374 go medium
480
Generated Behavioral Tests
100.0%
Best Score
Muse Spark 1.3 (max)

Results by Model

Hover for details · Click a model to open it · Drag to zoom the x-axis · Shift-drag to pan · The line marks the Pareto frontier (best score per cost)

Per-Model Results

25 runs
# Model Score Cost Calls
1 Muse Spark 1.3 (max) 100.0% $2.75 209 trace →
2 Muse Spark 1.3 (xhigh) 99.8% $1.33 267 trace →
3 Claude Opus 4.8 (xhigh) 99.8% $24.77 180 trace →
4 Claude Opus 5 (xhigh) 97.5% $16.09 127 trace →
5 Gemini 3.7 Flash 92.5% $1.46 139 trace →
6 GLM-5.2 87.3% $14.97 254 trace →
7 GPT-5.6 Sol (xhigh) 87.1% $2.81 24 trace →
8 GPT 5.5 (high) 81.9% $2.12 27 trace →
9 Gemini 3.5 Flash 81.0% $2.81 198 trace →
10 Claude Opus 4.6 80.8% $10.87 250 trace →
11 Muse Spark 1.2 (xhigh) 74.2% $0.47 111 trace →
12 GPT 5.5 (xhigh) 65.0% $6.65 85 trace →
13 Gemini 3.6 Flash 59.8% $7.64 259 trace →
14 GPT 5.5 58.8% $1.07 19 trace →
15 GPT-5.6 Sol 45.4% $0.57 10 trace →
16 Claude Sonnet 4.6 36.0% $11.82 325 trace →
17 Claude Opus 4.7 (xhigh) 25.8% $10.56 164 trace →
18 Claude Opus 4.7 24.6% $6.25 148 trace →
19 Gemini 3 Flash 23.3% $0.30 114 trace →
20 Claude Haiku 4.5 20.6% $1.12 153 trace →
21 GPT 5.4 9.2% $0.31 13 trace →
22 GPT 5 mini 8.1% $0.03 22 trace →
23 GPT 5.4 mini 1.0% $0.03 21 trace →
24 Muse Spark 1.1 (xhigh) 0.0% $5.19 691 trace →
25 Gemini 3.1 Pro 0.0% $0.96 83 trace →

Click a row to replay how that model rebuilt this program, or the model name to open its full run