Results
| Model | Effort | Harness | Score (%) | Date |
|---|---|---|---|---|
| GPT-6 Sol | default | codex-cli@1.4 | 64.0 | 2026-08-14 |
| GPT-6 Sol | max | codex-cli@1.4 | 71.2 | 2026-08-14 |
| GPT-6 Astra | default | codex-cli@1.4 | 58.3 | 2026-08-14 |
| Nimbus 3 (max effort) | codex-cli@1.4 | 66.6 | 2026-08-14 | |
| Nimbus 3 Mini | default | codex-cli@1.4 | 49.5 | 2026-08-14 |
Notes
Scores are the mean of five runs. Nimbus 3 at default effort was not submitted.