BrowseComp on BrowseComp (test)
67.67AccuracyArbor
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Arbor2026.06 | 67.67 | 22.34 | |
| Claude Code2026.06 | 53.33 | 8 | |
| Codex2026.06 | 50 | 4.67 | |
| Initial2026.06 | 45.33 | — |
Loading the SOTA2 catalog…
| Method | Links | ||
|---|---|---|---|
| Arbor2026.06 | 67.67 | 22.34 | |
| Claude Code2026.06 | 53.33 | 8 | |
| Codex2026.06 | 50 | 4.67 | |
| Initial2026.06 | 45.33 | — |