Terminal-Bench 2.0 Evaluation (test)
77.36Pass RateArbor
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Arbor2026.06 | 77.36 | 7.55 | |
| Codex2026.06 | 73.59 | 3.78 | |
| Claude Code2026.06 | 71.7 | 1.89 | |
| Initial2026.06 | 69.81 | — |
Loading the SOTA2 catalog…
| Method | Links | ||
|---|---|---|---|
| Arbor2026.06 | 77.36 | 7.55 | |
| Codex2026.06 | 73.59 | 3.78 | |
| Claude Code2026.06 | 71.7 | 1.89 | |
| Initial2026.06 | 69.81 | — |