List your company

Software engineering agent task completion on SWE-Bench Verified

78.2AccuracyLLM-as-a-Verifier
75.70476.3527777.648Jul 6, 2026

Evaluation results

MethodLinks
LLM-as-a-Verifier2026.0778.2——
Opus 4.5Baseline Rank=#12026.0776.8——
G3 FlashBaseline Rank=#22026.0775.8——
M2.5Baseline Rank=#32026.0775.8——
Candidate Pool2026.07—76.184.4