pemistahl/grex
A command-line tool and Rust library with Python bindings for generating regular expressions from user-provided test cases
1,312
Generated Behavioral Tests
98.1%
Best Score
Claude Opus 4.8 (xhigh)
Results by Model
Hover a point for details · The line marks the Pareto frontier (best score per cost) · Click a point to see model details
Per-Model Results
19 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 | |
GPT 5.5 (high) | 88.7% | $6.81 | 57 | trace → |
| 5 |  |
Gemini 3.5 Flash | 84.4% | $6.38 | 205 | trace → |
| 6 | |
Claude Opus 4.7 (xhigh) | 82.1% | $31.59 | 231 | trace → |
| 7 | |
Gemini 3.6 Flash | 81.8% | $46.97 | 435 | trace → |
| 8 | |
GPT 5.6 Sol (medium) | 81.5% | $1.37 | 24 | trace → |
| 9 | |
Claude Opus 4.7 | 73.9% | $7.62 | 153 | trace → |
| 10 |  |
GLM-5.2 | 70.9% | $40.95 | 292 | trace → |
| 11 | |
GPT 5.4 | 66.5% | $0.32 | 7 | trace → |
| 12 | |
Claude Opus 4.6 | 66.0% | $16.93 | 266 | trace → |
| 13 | |
Gemini 3.1 Pro | 61.4% | $1.29 | 159 | trace → |
| 14 | |
GPT 5.5 | 58.1% | $1.56 | 22 | trace → |
| 15 | |
Claude Sonnet 4.6 | 56.3% | $48.82 | 743 | trace → |
| 16 | |
Gemini 3 Flash | 52.8% | $0.23 | 81 | trace → |
| 17 | |
GPT 5 mini | 34.7% | $0.08 | 36 | trace → |
| 18 | |
GPT 5.4 mini | 21.6% | $0.05 | 10 | trace → |
| 19 | |
Claude Haiku 4.5 | 0.0% | $1.02 | 104 | trace → |
Click a row to replay how that model rebuilt this program, or the model name to open its full run