List your company

Code Security Vulnerability Repair on LiveCVEBench (LCB) (test)

41.27Success RateClaude Opus 4.5
3.850813.565423.2832.9946Feb 3, 2026

Evaluation results

MethodLinks
Claude Opus 4.5Base Model=Claude Opus 4.52026.0241.27
Qwen3-32B + CVE-Factory (4k)Fine-tuning Data Source=CVE-Factory, Training Data Scale (Trajectories)=4k, Base Model=Qwen3-32B2026.0235.79
Claude Sonnet 4.5Base Model=Claude Sonnet 4.52026.0234.39
Qwen3-32B + CVE-Factory (3k)Fine-tuning Data Source=CVE-Factory, Training Data Scale (Trajectories)=3k, Base Model=Qwen3-32B2026.0231.05
MiniMax-M2Base Model=MiniMax-M22026.0224.87
Qwen3-32B + SETAFine-tuning Data Source=SETA, Training Data Scale (Trajectories)=4k, Base Model=Qwen3-32B2026.0221.69
Claude Sonnet 4Base Model=Claude Sonnet 42026.0220.11
Qwen 3 Coder 480BBase Model=Qwen 3 Coder 480B2026.0219.58
Qwen3-Coder-30BBase Model=Qwen3-Coder-30B2026.0210.58
Qwen3-32BBase Model=Qwen3-32B2026.025.29