Notealy / Aegis research
Research reports.
Every published evaluation of Aegis, with its setup, results, error analysis, and limitations. Results are published as measured, including the ones that show what the model cannot do yet.
Report 001
Baseline evaluation of Aegis-1
We measured the Aegis-1 development build and its pinned reference baseline on 24 held-out security reasoning cases, across detection, classification, and end-to-end analysis. Both models find most vulnerable code but also flag most secure code, and neither assigns a correct CWE category. These results set the training targets for the first fine-tuned version.
Notealy
Follow the workas it happens.
Milestones, benchmark definitions, and methodology live on the Aegis progress page.