Versions

1 version, plus 3 variants and configurations · Metric: CWE-bench Programmatic Pass@1

Versions of CWE-bench
VersionReleasedModelsResults
CWE-bench 1Latest
3 variants of 1
Variants of CWE-bench 1
VariantConfigurationReleasedModelsResults
CWE-bench 1 — Claude CodeClaude Code harnessSeptember 25, 202622
CWE-bench 1 — CodexCodex harnessSeptember 25, 202622
CWE-bench 1 — OpenCodeOpenCode harnessSeptember 25, 202644
September 25, 202611

Recently reported results: CWE-bench 1

Latest version: CWE-bench 1

Recently reported results for CWE-bench 1
ModelScoreSource
Gemini 4 Argon (highest thinking settings)68.0%deepmind.google for Gemini 4 Argon (highest thinking settings) on CWE-bench 1 (opens in a new tab)CWE-bench Programmatic Pass@1Reported Evaluation methodologyGoogle DeepMindSelf-reportedCollinear AIRecord for Gemini 4 Argon on CWE-bench 1