RustMizan - Vulnerability Detection: leaderboard

Metric: Vulnerability classification accuracy (%; whether each of the 173 compilable crate-, file- and function-level Rust variants of 42 RustSec memory-safety CVEs, unmutated, each analysed by a ReAct agent with shell access in an Inspect sandbox that returns a structured JSON verdict is correctly called vulnerable or not, 95 of the 173 being vulnerable; higher is better). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.

Top models

#ModelScore
1Claude Sonnet 4.664.7
2GPT-5.464.2
3Gemini 3.1 Pro (Preview)57.2
4Qwen 3.6 Plus56.1

Interactive version: theaggregate.ai/benchmark?slug=rustmizan-vulnerability-detection · How It Works · Data refreshed daily, snapshot 2026-09-29.