← Back to leaderboard  ·  Show all task instances

rs/curlie

The power of curl, the ease of use of httpie.

3,637 go easy
693
Generated Behavioral Tests
97.1%
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) 97.1% $3.34 247 trace →
2 Claude Opus 5 (xhigh) 95.8% $18.53 146 trace →
3 Claude Opus 4.7 (xhigh) 92.9% $18.14 157 trace →
4 Claude Opus 4.8 (xhigh) 92.9% $14.69 109 trace →
5 GPT 5.5 (xhigh) 92.8% $7.88 81 trace →
6 Gemini 3.7 Flash 92.5% $1.84 142 trace →
7 Muse Spark 1.3 (xhigh) 91.2% $1.97 258 trace →
8 Muse Spark 1.1 (xhigh) 89.9% $0.52 240 trace →
9 Claude Opus 4.6 89.3% $12.49 243 trace →
10 GLM-5.2 88.9% $22.71 261 trace →
11 Claude Sonnet 4.6 88.3% $25.25 518 trace →
12 GPT 5.5 (high) 87.7% $3.25 38 trace →
13 Muse Spark 1.2 (xhigh) 87.0% $1.06 139 trace →
14 Gemini 3.6 Flash 85.1% $4.78 122 trace →
15 Claude Opus 4.7 84.8% $15.70 247 trace →
16 GPT 5.5 83.1% $1.21 23 trace →
17 Gemini 3.5 Flash 80.4% $2.41 99 trace →
18 Gemini 3.1 Pro 79.8% $1.62 127 trace →
19 GPT-5.6 Sol (xhigh) 76.3% $5.88 43 trace →
20 Gemini 3 Flash 71.1% $0.22 85 trace →
21 GPT-5.6 Sol 70.0% $1.17 19 trace →
22 GPT 5.4 54.0% $0.27 10 trace →
23 Claude Haiku 4.5 43.0% $0.59 118 trace →
24 GPT 5.4 mini 11.3% $0.02 9 trace →
25 GPT 5 mini 0.0% $0.03 16 trace →

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