← Back to leaderboard  ·  Show all task instances

bootandy/dust

A more intuitive version of du in rust

11,609 rs medium
584
Generated Behavioral Tests
91.8%
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) 91.8% $71.18 297 trace →
2 GPT 5.5 (xhigh) 83.4% $12.40 89 trace →
3 Muse Spark 1.3 (max) 82.2% $14.65 547 trace →
4 Claude Opus 4.8 (xhigh) 81.8% $53.90 272 trace →
5 Muse Spark 1.3 (xhigh) 80.8% $2.27 258 trace →
6 Claude Opus 4.7 (xhigh) 77.9% $42.44 255 trace →
7 GPT 5.5 (high) 77.1% $3.25 28 trace →
8 Gemini 3.7 Flash 73.5% $3.72 151 trace →
9 Claude Opus 4.7 70.9% $12.12 171 trace →
10 GPT-5.6 Sol (xhigh) 70.0% $6.44 37 trace →
11 Claude Opus 4.6 70.0% $15.78 245 trace →
12 Claude Sonnet 4.6 68.7% $45.15 499 trace →
13 GPT-5.6 Sol 65.2% $1.65 15 trace →
14 Muse Spark 1.1 (xhigh) 64.0% $0.66 228 trace →
15 Muse Spark 1.2 (xhigh) 63.9% $1.62 356 trace →
16 Gemini 3.5 Flash 63.2% $6.25 139 trace →
17 GPT 5.5 60.6% $1.19 14 trace →
18 GLM-5.2 60.1% $62.20 427 trace →
19 Gemini 3.6 Flash 58.2% $3.30 128 trace →
20 GPT 5.4 52.2% $0.46 11 trace →
21 Gemini 3.1 Pro 45.7% $1.32 88 trace →
22 Claude Haiku 4.5 42.8% $1.46 164 trace →
23 Gemini 3 Flash 28.1% $0.21 53 trace →
24 GPT 5.4 mini 18.3% $0.04 8 trace →
25 GPT 5 mini 2.1% $0.05 18 trace →

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