Semgrep IDOR - Multimodal F1: leaderboard
Semgrep researcher-reviewed benchmark of insecure direct object reference detection in representative production applications over ten runs, using the Semgrep Multimodal security workflow and reporting F1.
Metric: F1 score (%). Source: semgrep.dev. Saturation forecast: Around October 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 61.7 | #19 |
| 2 | GPT-5.6 Luna | 61.6 | #68 |
| 3 | GPT-5.6 Terra | 59.3 | #41 |
| 4 | GLM-5.2 | 58.1 | #73 |
| 5 | GPT-5.5 | 48.1 | #31 |
| 6 | Claude Opus 4.8 | 44.8 | #37 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=semgrep-idor-multimodal-f1 · How It Works · Data refreshed daily, snapshot 2026-10-20.