List your company

Code Security Vulnerability Repair on PatchEval (PE) (test)

32.08Success RateClaude Opus 4.5
4.603211.736618.8726.0034Feb 3, 2026

Evaluation results

MethodLinks
Claude Opus 4.5Base Model=Claude Opus 4.52026.0232.08
Claude Sonnet 4.5Base Model=Claude Sonnet 4.52026.0228.77
Qwen3-32B + CVE-Factory (4k)Fine-tuning Data Source=CVE-Factory, Training Data Scale (Trajectories)=4k, Base Model=Qwen3-32B2026.0223.58
Claude Sonnet 4Base Model=Claude Sonnet 42026.0222.64
Qwen3-32B + CVE-Factory (3k)Fine-tuning Data Source=CVE-Factory, Training Data Scale (Trajectories)=3k, Base Model=Qwen3-32B2026.0219.81
Qwen 3 Coder 480BBase Model=Qwen 3 Coder 480B2026.0219.34
MiniMax-M2Base Model=MiniMax-M22026.0219.34
Qwen3-32B + SETAFine-tuning Data Source=SETA, Training Data Scale (Trajectories)=4k, Base Model=Qwen3-32B2026.0214.62
Qwen3-Coder-30BBase Model=Qwen3-Coder-30B2026.029.91
Qwen3-32BBase Model=Qwen3-32B2026.025.66