← Back to leaderboard  ·  Show all task instances

sheepla/pingu

🐧ping command but with pingu

2,087 go easy
378
Generated Behavioral Tests
99.7%
Best Score
Muse Spark 1.3 (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 Muse Spark 1.3 (xhigh) 99.7% $0.45 120 trace →
2 Claude Opus 4.7 (xhigh) 98.9% $5.81 109 trace →
3 Muse Spark 1.2 (xhigh) 98.4% $2.31 187 trace →
4 Gemini 3.7 Flash 97.9% $0.75 55 trace →
5 Claude Opus 4.8 (xhigh) 97.6% $8.15 77 trace →
6 Claude Opus 4.7 96.8% $3.36 104 trace →
7 Muse Spark 1.3 (max) 96.6% $2.07 190 trace →
8 GPT-5.6 Sol (xhigh) 96.3% $1.97 19 trace →
9 GPT 5.5 (xhigh) 96.3% $4.39 45 trace →
10 GPT 5.5 (high) 96.0% $2.73 35 trace →
11 GLM-5.2 95.4% $9.15 110 trace →
12 Claude Opus 5 (xhigh) 95.0% $8.03 74 trace →
13 GPT-5.6 Sol 94.7% $0.88 16 trace →
14 Claude Opus 4.6 94.2% $6.93 165 trace →
15 Gemini 3.6 Flash 93.1% $3.32 101 trace →
16 Claude Sonnet 4.6 92.3% $3.78 175 trace →
17 Gemini 3.5 Flash 91.3% $2.95 96 trace →
18 GPT 5.5 87.8% $0.93 19 trace →
19 Claude Haiku 4.5 73.5% $0.48 87 trace →
20 GPT 5 mini 66.1% $0.02 18 trace →
21 GPT 5.4 61.9% $0.22 10 trace →
22 Gemini 3 Flash 58.7% $0.30 59 trace →
23 GPT 5.4 mini 17.5% $0.01 9 trace →
24 Muse Spark 1.1 (xhigh) 0.5% $0.29 136 trace →
25 Gemini 3.1 Pro 0.0% $1.11 48 trace →

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