← Back to leaderboard  ·  Show all task instances

Canop/rhit

A nginx log explorer

1,006 rs medium
817
Generated Behavioral Tests
88.4%
Best Score
Muse Spark 1.3 (max)

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 (max) 88.4% $4.20 246 trace →
2 Claude Opus 4.8 (xhigh) 81.8% $22.51 141 trace →
3 GPT-5.6 Sol (xhigh) 78.9% $7.95 35 trace →
4 Muse Spark 1.2 (xhigh) 74.9% $2.81 362 trace →
5 Gemini 3.6 Flash 74.7% $6.20 155 trace →
6 GLM-5.2 73.7% $35.78 217 trace →
7 GPT 5.5 (xhigh) 73.2% $12.36 78 trace →
8 Gemini 3.7 Flash 73.2% $6.51 204 trace →
9 Muse Spark 1.1 (xhigh) 71.7% $0.29 131 trace →
10 GPT 5.5 (high) 70.1% $4.62 43 trace →
11 GPT-5.6 Sol 69.3% $1.02 11 trace →
12 Gemini 3.5 Flash 67.9% $11.46 193 trace →
13 Muse Spark 1.3 (xhigh) 66.8% $7.37 541 trace →
14 Claude Opus 4.7 (xhigh) 57.5% $1.65 34 trace →
15 Claude Opus 4.7 53.2% $1.18 34 trace →
16 Claude Sonnet 4.6 52.5% $40.64 458 trace →
17 Claude Opus 4.6 52.4% $15.70 230 trace →
18 GPT 5.5 44.8% $1.74 17 trace →
19 GPT 5.4 43.2% $0.23 7 trace →
20 Gemini 3 Flash 38.4% $0.40 68 trace →
21 Gemini 3.1 Pro 31.1% $3.17 135 trace →
22 Claude Haiku 4.5 28.5% $1.08 134 trace →
23 GPT 5 mini 7.8% $0.02 15 trace →
24 GPT 5.4 mini 1.3% $0.07 17 trace →
25 Claude Opus 5 (xhigh) 0.0% $8.30 63 trace →

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