Results

ModelEffortHarnessScore (%)Date
GPT-6 Soldefaultcodex-cli@1.464.02026-08-14
GPT-6 Solmaxcodex-cli@1.471.22026-08-14
GPT-6 Astradefaultcodex-cli@1.458.32026-08-14
Nimbus 3 (max effort)codex-cli@1.466.62026-08-14
Nimbus 3 Minidefaultcodex-cli@1.449.52026-08-14

Notes

Scores are the mean of five runs. Nimbus 3 at default effort was not submitted.