run-0106
capability
citable URL: https://halobench.com/records/run-0106/ — this address never moves; the anchor /records/#run-0106 keeps resolving
date 2026-08-11 · window 2026-08-11T12:01:45Z → 2026-08-11T12:22:32Z
suite tau2-bench-airline@668d3bc · config cfg-0031
depth 32,768 tokens · scoring deterministic
Metrics
| metric | value |
|---|---|
| tasks_passed | 9 |
| tasks_total | 9 |
| mean_score | 1 |
turns: total 162 · median 15 · p90 29 · max 29 — machine-written trailer, parsed at build time
Guard chain
not a performance run — no guard required
Energy
metered window: eng-0044
Per-task trace — the full record, no truncation
| task | reward | turns | seconds |
|---|---|---|---|
| 0 | 1 | 20 | 118 |
| 1 | 1 | 24 | 122 |
| 2 | 1 | 29 | 211 |
| 3 | 1 | 14 | 73 |
| 4 | 1 | 12 | 84 |
| 5 | 1 | 15 | 123 |
| 6 | 1 | 10 | 89 |
| 8 | 1 | 26 | 197 |
| 9 | 1 | 12 | 138 |
reward null = infrastructure error, excluded from every paired computation
Cited by — computed at build time, never stored
claims clm-0046