← Back to leaderboard  ·  Show all task instances

pemistahl/grex

A command-line tool and Rust library with Python bindings for generating regular expressions from user-provided test cases

8,103 rs medium
1,312
Generated Behavioral Tests
98.1%
Best Score
Claude Opus 4.8 (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 4.8 (xhigh) 98.1% $17.62 125 trace →
2 GPT-5.6 Sol (xhigh) 97.4% $13.41 78 trace →
3 GPT 5.5 (xhigh) 97.2% $9.20 71 trace →
4 Muse Spark 1.3 (max) 94.1% $5.89 296 trace →
5 Gemini 3.7 Flash 91.8% $9.95 264 trace →
6 Muse Spark 1.1 (xhigh) 88.8% $0.54 222 trace →
7 GPT 5.5 (high) 88.7% $6.81 57 trace →
8 Gemini 3.5 Flash 84.4% $6.38 205 trace →
9 Muse Spark 1.2 (xhigh) 84.2% $8.73 486 trace →
10 Claude Opus 4.7 (xhigh) 82.1% $31.59 231 trace →
11 Gemini 3.6 Flash 81.8% $46.97 435 trace →
12 GPT-5.6 Sol 81.5% $1.37 24 trace →
13 Claude Opus 4.7 73.9% $7.62 153 trace →
14 GLM-5.2 70.9% $40.95 292 trace →
15 GPT 5.4 66.5% $0.32 7 trace →
16 Claude Opus 4.6 66.0% $16.93 266 trace →
17 Gemini 3.1 Pro 61.4% $1.29 159 trace →
18 GPT 5.5 58.1% $1.56 22 trace →
19 Claude Sonnet 4.6 56.3% $48.82 743 trace →
20 Gemini 3 Flash 52.8% $0.23 81 trace →
21 GPT 5 mini 34.7% $0.08 36 trace →
22 GPT 5.4 mini 21.6% $0.05 10 trace →
23 Claude Opus 5 (xhigh) 0.0% $17.74 116 trace →
24 Claude Haiku 4.5 0.0% $1.02 104 trace →
25 Muse Spark 1.3 (xhigh) n/a $7.03 505 trace →

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