LLM Leaderboard for Code Quality & Security

Independent analysis of code bugs, vulnerabilities, and code smells for leading LLMs


Leaderboard

Models ranked by pass rate. Additional metrics: issue density, cognitive complexity, and static analysis findings from SonarQube Cloud.

Want to add a model or license data? Contact us (opens in a new tab)

LLM leaderboard. Column headers with buttons sort the table.
Model Vulnerabilities Issues / MLOC Bugs Issues / MLOC Code smells Issues / MLOC

Vulnerability Distribution

Vulnerability issues per MLOC broken down by severity — lower is better. Showing up to 25 best-ranked models matching your filters (lowest total vulnerability density).

  • Blocker
  • Critical
  • Major
  • Minor

Issues / MLOC (lower is better)

Bugs Distribution

Bug issues per MLOC broken down by severity — lower is better. Showing up to 25 best-ranked models matching your filters (lowest total bug density).

  • Blocker
  • Critical
  • Major
  • Minor

Issues / MLOC (lower is better)

Code Smells Distribution

Code smell issues per MLOC broken down by severity — lower is better. Showing up to 25 best-ranked models matching your filters (lowest total code smell density).

  • Blocker
  • Critical
  • Major
  • Minor

Issues / MLOC (lower is better)