Dafny Code Synthesis on Dafny Benchmark Overall Vericoding (aggregate)
82.2Pass RateModel union
Evaluation Results
| Method | Links | |
|---|---|---|
| Model unionAttempts per task=52026.05 | 82.2 | |
| Claude Opus 4.1Attempts per task=52026.05 | 67.5 | |
| GPT-5 miniAttempts per task=52026.05 | 66.9 | |
| Gemini 2.5 FlashAttempts per task=52026.05 | 38.2 |