Compare systems
Compare systems
Pick up to five systems from any leaderboard and see their pass rates side by side.
Side by side3 of 18 systems on the Vulnerability repair · Patch leaderboard
Each bar is the share of vulnerabilities where the system's patch passed the grader. The reasoning effort for these runs is not recorded.
RankSystemPass rate
1
Codex GPT-5.5
86.3%80.5–90.61
OpenCode GPT-5.5
85.3%78.3–90.81
OpenCode GPT-5.3 Codex
84.0%78.0–88.6Leaderboard: Vulnerability repair, Patch · newest run 30 April 2026 · Open the leaderboard