← Back to leaderboard  ·  Show all task instances

madler/pigz

A parallel implementation of gzip for modern multi-processor, multi-core machines.

2,924 c medium
831
Generated Behavioral Tests
98.6%
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

24 runs
# Model Score Cost Calls
1 Claude Opus 5 (xhigh) 98.6% $24.31 181 trace →
2 Muse Spark 1.3 (max) 98.1% $4.65 327 trace →
3 Muse Spark 1.3 (xhigh) 97.1% $1.85 258 trace →
4 GPT 5.5 (xhigh) 94.9% $10.57 91 trace →
5 GPT-5.6 Sol (xhigh) 93.6% $4.26 36 trace →
6 Claude Opus 4.7 (xhigh) 92.2% $8.06 152 trace →
7 Muse Spark 1.2 (xhigh) 91.8% $1.02 129 trace →
8 GPT-5.6 Sol 90.6% $1.02 22 trace →
9 GPT 5.5 (high) 89.4% $2.86 38 trace →
10 Gemini 3.5 Flash 85.1% $3.72 161 trace →
11 GPT 5.5 84.2% $1.03 17 trace →
12 Claude Opus 4.7 83.2% $4.53 149 trace →
13 GLM-5.2 80.6% $39.11 213 trace →
14 GPT 5.4 79.1% $0.23 10 trace →
15 Claude Sonnet 4.6 74.8% $11.64 343 trace →
16 Claude Haiku 4.5 73.4% $0.47 84 trace →
17 Gemini 3.1 Pro 66.4% $1.10 88 trace →
18 Gemini 3.7 Flash 50.1% $1.81 112 trace →
19 Claude Opus 4.8 (xhigh) 47.9% $21.11 140 trace →
20 Gemini 3 Flash 29.6% $0.14 37 trace →
21 GPT 5 mini 10.6% $0.01 8 trace →
22 Muse Spark 1.1 (xhigh) 2.3% $0.16 92 trace →
23 Claude Opus 4.6 0.0% $1.63 215 trace →
24 GPT 5.4 mini 0.0% $0.02 13 trace →

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