← Back to leaderboard  ·  Show all task instances

hooklift/gowsdl

WSDL2Go code generation as well as its SOAP proxy

1,219 go medium
391
Generated Behavioral Tests
96.2%
Best Score
Claude Opus 5 (xhigh)

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 Claude Opus 5 (xhigh) 96.2% $35.79 215 trace →
2 Claude Opus 4.8 (xhigh) 94.9% $40.33 233 trace →
3 Muse Spark 1.3 (xhigh) 93.9% $4.17 449 trace →
4 GPT-5.6 Sol (xhigh) 93.6% $11.11 79 trace →
5 Muse Spark 1.3 (max) 92.1% $5.48 295 trace →
6 GPT 5.5 (xhigh) 90.8% $17.46 117 trace →
7 Gemini 3.7 Flash 89.5% $3.51 181 trace →
8 Claude Opus 4.7 (xhigh) 89.5% $21.55 224 trace →
9 Claude Opus 4.7 86.4% $19.71 260 trace →
10 GPT 5.5 (high) 85.7% $7.43 73 trace →
11 Muse Spark 1.1 (xhigh) 85.4% $1.70 264 trace →
12 GLM-5.2 82.9% $36.21 361 trace →
13 Muse Spark 1.2 (xhigh) 81.6% $0.18 151 trace →
14 GPT 5.5 80.6% $1.60 28 trace →
15 GPT-5.6 Sol 78.8% $2.24 24 trace →
16 Gemini 3.5 Flash 78.8% $4.53 166 trace →
17 Gemini 3.1 Pro 69.3% $1.04 82 trace →
18 Claude Haiku 4.5 63.2% $1.02 155 trace →
19 Claude Sonnet 4.6 62.4% $60.18 801 trace →
20 Claude Opus 4.6 61.6% $20.18 296 trace →
21 Gemini 3.6 Flash 31.7% $5.65 153 trace →
22 Gemini 3 Flash 30.7% $0.27 62 trace →
23 GPT 5.4 mini 26.6% $0.02 13 trace →
24 GPT 5.4 11.8% $0.38 10 trace →
25 GPT 5 mini 11.5% $0.01 9 trace →

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