Evaluation
Window Details
Coming soon. Current rows are simulated and show the planned support layer behind Boundary Results.
| model | dataset_window | result | closed_choice | idk_enabled | evaluation_date | Actions |
|---|---|---|---|---|---|---|
| DeepSeek V4 Flash | 2026-05 | Inconclusive | 18.0% | 11.0% | 2026-06-19 | Open |
| Frontier Model Alpha | 2024-12 | High look-ahead risk | 44.0% | 32.0% | 2026-06-19 | Open |
Detailed methodology copy and official records will be published after the release policy is finalized.