Software engineering agent task completion on SWE-Bench Verified
78.2AccuracyLLM-as-a-Verifier
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLM-as-a-Verifier2026.07 | 78.2 | — | — | |
| Opus 4.5Baseline Rank=#12026.07 | 76.8 | — | — | |
| G3 FlashBaseline Rank=#22026.07 | 75.8 | — | — | |
| M2.5Baseline Rank=#32026.07 | 75.8 | — | — | |
| Candidate Pool2026.07 | — | 76.1 | 84.4 |