← Back to leaderboard  ·  Show all task instances

ogham/dog

A command-line DNS client.

6,640 rs medium
1,300
Generated Behavioral Tests
99.6%
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) 99.6% $63.51 350 trace →
2 Muse Spark 1.3 (max) 99.0% $3.80 255 trace →
3 Muse Spark 1.3 (xhigh) 98.1% $1.57 215 trace →
4 GPT-5.6 Sol (xhigh) 97.9% $5.77 41 trace →
5 Claude Opus 4.8 (xhigh) 96.3% $17.64 145 trace →
6 GPT 5.5 (xhigh) 93.9% $8.10 68 trace →
7 Gemini 3.7 Flash 93.1% $1.88 102 trace →
8 Muse Spark 1.2 (xhigh) 91.3% $5.45 266 trace →
9 GPT 5.5 (high) 91.2% $3.43 41 trace →
10 Gemini 3.6 Flash 90.8% $4.40 162 trace →
11 Claude Opus 4.7 (xhigh) 90.5% $15.42 222 trace →
12 Claude Opus 4.6 84.2% $11.72 273 trace →
13 Claude Opus 4.7 80.8% $5.47 144 trace →
14 GLM-5.2 78.8% $27.76 219 trace →
15 Claude Sonnet 4.6 76.5% $7.18 249 trace →
16 GPT 5.5 75.4% $1.32 20 trace →
17 GPT-5.6 Sol 72.9% $0.42 11 trace →
18 Gemini 3.1 Pro 70.8% $1.14 99 trace →
19 Gemini 3 Flash 69.0% $0.30 94 trace →
20 Claude Haiku 4.5 51.5% $0.74 112 trace →
21 GPT 5.4 28.8% $0.17 7 trace →
22 GPT 5.4 mini 17.8% $0.03 7 trace →
23 GPT 5 mini 10.2% $0.02 17 trace →
24 Gemini 3.5 Flash 8.8% $5.66 164 trace →
25 Muse Spark 1.1 (xhigh) 0.0% $2.28 893 trace →

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