run-0108
capability
citable URL: https://halobench.com/records/run-0108/ — this address never moves; the anchor /records/#run-0108 keeps resolving
date 2026-08-11 · window 2026-08-11T16:31:26Z → 2026-08-11T17:55:14Z
suite tau2-bench-airline@668d3bc · config cfg-0027
depth 32,768 tokens · scoring deterministic
Metrics
| metric | value |
|---|---|
| tasks_passed | 7 |
| tasks_total | 12 |
| mean_score | 0.5833 |
turns: total 630 · median 38 · p90 84 · max 201 — machine-written trailer, parsed at build time
Guard chain
not a performance run — no guard required
Energy
metered window: eng-0062
Per-task trace — the full record, no truncation
| task | reward | turns | seconds |
|---|---|---|---|
| 0 | 1 | 24 | 199 |
| 1 | 1 | 36 | 191 |
| 2 | 0 | 84 | 538 |
| 3 | 1 | 16 | 94 |
| 4 | 1 | 38 | 265 |
| 5 | 1 | 24 | 206 |
| 6 | 0 | 46 | 1081 |
| 8 | 1 | 22 | 188 |
| 9 | 0 | 201 | 805 |
| 10 | 0 | 54 | 750 |
| 11 | 1 | 34 | 337 |
| 12 | 0 | 51 | 326 |
reward null = infrastructure error, excluded from every paired computation
Cited by — computed at build time, never stored
claims clm-0047
model pages nemotron3-super