BMS

Start typing to search the publication.

Shareable benchmark figure

Codex scored higher on the reconstructed finding set

Recall, manual precision, and F1 for submitted audit findings. Higher is better.

Figure 02 · Audit quality

Codex scored higher on the reconstructed finding set

GPT-5.6 SolGrok 4.6 Build
Figure 02Recall, manual precision, and F1 for submitted audit findings. Higher is better.

One run per model. Harnesses differed. Manually reconstructed ground truth under a closed-world policy; unmatched headings are not automatically invalid. Accuracy is undefined.

BMSbms.kyokasuigetsu.xyz
Share figure
Static image: https://bms.kyokasuigetsu.xyz/charts/quality-metrics.png

Read the full benchmark