KSXGitHub/parallel-disk-usage

Highly parallelized, blazing fast directory tree analyzer

689 rs medium
531
Generated Behavioral Tests
99.6%
Best Score
Claude Opus 5 (xhigh)

Hover a point for details · The line marks the Pareto frontier (best score per cost) · Click a point to see model details

21 runs
# Model Score Cost Calls
1 Claude Opus 5 (xhigh) 99.6% $35.38 213 trace →
2 Gemini 3.7 Flash 96.0% $2.77 171 trace →
3 GPT 5.5 (xhigh) 95.9% $6.67 57 trace →
4 GPT-5.6 Sol (xhigh) 91.5% $3.52 22 trace →
5 Gemini 3.6 Flash 87.6% $13.76 236 trace →
6 Claude Opus 4.6 86.1% $18.23 273 trace →
7 Claude Sonnet 4.6 85.1% $29.37 456 trace →
8 Claude Opus 4.7 (xhigh) 83.4% $13.81 153 trace →
9 GPT 5.5 (high) 82.7% $3.38 32 trace →
10 GPT-5.6 Sol 81.7% $1.02 13 trace →
11 Claude Opus 4.7 76.5% $4.66 99 trace →
12 GPT 5.5 73.6% $1.01 14 trace →
13 Gemini 3.5 Flash 66.5% $5.30 124 trace →
14 GPT 5.4 63.3% $0.60 14 trace →
15 Gemini 3 Flash 57.3% $0.42 95 trace →
16 Claude Haiku 4.5 55.4% $1.02 140 trace →
17 Gemini 3.1 Pro 54.4% $2.34 129 trace →
18 GPT 5 mini 13.7% $0.02 13 trace →
19 Claude Opus 4.8 (xhigh) 0.0% $24.77 162 trace →
20 GLM-5.2 0.0% $51.89 291 trace →
21 GPT 5.4 mini 0.0% $0.03 11 trace →

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