List your company

Vulnerability Attack Performance on E2E (dev)

100ASR (Semgrep) - E-commerceGemini-3 Pro
-4235077Feb 2, 2026

Evaluation results

MethodLinks
Gemini-3 ProEvaluation Protocol=Held-out (target-domain)2026.0210010010010033.3333.3333.3333.331001001001001007510075100100100100
Gemini-3 FlashEvaluation Protocol=Held-out (target-domain)2026.021001001001001005010050100100100100100751007510066.6710066.67
GrokEvaluation Protocol=Held-out (target-domain)2026.02100100100100—50—50010001001005010050100100100100
GPT-5.2Evaluation Protocol=Held-out (target-domain)2026.020——010033.3310033.330100010010075100751005010050
Claude-4.5 OpusEvaluation Protocol=Held-out (target-domain)2026.0201001000501005010001000100100100100100050050
ComposerEvaluation Protocol=Held-out (target-domain)2026.020——0505050500100010010050100500000