SecIT Bench (Claude Code): leaderboard
The SecIT Bench scenario set run inside the Claude Code agent product rather than a plain model harness. Scored identically to the Pydantic AI board, and kept separate because the scaffold is part of what is being measured: the same model moves by up to three points between harnesses.
Metric: Accuracy (%). Source: secitbench.cribl.io. Status: years away from saturation. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 | 78.64 |
| 2 | Claude Opus 4.8 | 67.14 |
| 3 | Claude Sonnet 5 | 65.93 |
Interactive version: theaggregate.ai/benchmark?slug=secit-bench-claude-code · How It Works · Data refreshed daily, snapshot 2026-09-05.