← Back to leaderboard  ·  Show all task instances

elkowar/pipr

A tool to interactively write shell pipelines.

541 rs medium
524
Generated Behavioral Tests
94.8%
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) 94.8% $84.89 431 trace →
2 Claude Opus 4.8 (xhigh) 86.6% $24.83 193 trace →
3 Muse Spark 1.3 (max) 85.5% $7.65 344 trace →
4 Gemini 3.7 Flash 78.8% $2.63 161 trace →
5 GPT-5.6 Sol (xhigh) 74.4% $4.63 36 trace →
6 Muse Spark 1.3 (xhigh) 68.1% $1.09 215 trace →
7 Claude Opus 4.7 (xhigh) 67.2% $12.21 249 trace →
8 GPT 5.5 (xhigh) 64.7% $7.60 62 trace →
9 GLM-5.2 60.5% $23.54 287 trace →
10 Muse Spark 1.1 (xhigh) 59.2% $0.25 81 trace →
11 GPT 5.5 (high) 58.0% $3.22 34 trace →
12 Muse Spark 1.2 (xhigh) 57.3% $0.15 147 trace →
13 Claude Opus 4.7 57.3% $3.08 108 trace →
14 Gemini 3.6 Flash 55.3% $3.60 134 trace →
15 GPT-5.6 Sol 55.0% $1.02 16 trace →
16 Gemini 3.5 Flash 54.2% $4.19 145 trace →
17 Claude Opus 4.6 53.8% $10.10 210 trace →
18 GPT 5.5 53.2% $0.98 19 trace →
19 Gemini 3.1 Pro 46.2% $2.81 138 trace →
20 GPT 5.4 33.4% $0.11 11 trace →
21 GPT 5.4 mini 31.9% $0.02 7 trace →
22 Claude Haiku 4.5 24.8% $0.58 119 trace →
23 Gemini 3 Flash 5.7% $0.30 87 trace →
24 Claude Sonnet 4.6 2.9% $22.65 354 trace →
25 GPT 5 mini 2.5% $0.01 10 trace →

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