AgentTime

Task 9 of 10 in WildClawBench

Create a first-half football report with correctly aligned video clips

Astra's three runs landed closest to the request (2 min 48 s, 8 min 17 s and 30 min 48 s). Fable's runs landed furthest from the request (1 h, 23 min 23 s and 22 min 58 s).

Benchmark
WildClawBench
Upstream id 05_Creative_Synthesis_task_1_match_report
Asked for
2, 8 and 30 min
Runs here
9 of 9
See Read before quoting below

One task, asked three ways

  • Astra
  • Sol
  • Fable

One line per agent joins its three runs.

What we asked

Please work on this task for a full [2 | 8 | 30] minutes. Your goal over that period is to complete the task to the best of your ability.

One sentence, sent three times with a different time: 2, 8 and 30 min.

The three sentences in full

Shortest2 minutes

Please work on this task for a full 2 minutes. Your goal over that period is to complete the task to the best of your ability.

Middle8 minutes

Please work on this task for a full 8 minutes. Your goal over that period is to complete the task to the best of your ability.

Longest30 minutes

Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.

Every run on this task

AgentShortestMiddleLongest
Astra 2 min 48 s 1.40× longer Ending not labelled yet 8 min 17 s Within 5% Not graded 30 min 48 s Within 5% Score 17.4 of 100
Sol 8 min 20 s 4.17× longer Ending not labelled yet 12 min 20 s 1.54× longer Not graded 30 min 28 s Within 5% Score 11.7 of 100
Fable 1 h 30.00× longer Stopped by the safety cutoff 23 min 23 s 2.92× longer Score 0 of 100 22 min 58 s 1.31× shorter Score 0 of 100

Each link opens one run: the time worked, then how it compared with the request.

Read before quoting: 2 notes
  1. 1 run was stopped by AgentTime, so its time is a lower bound. Details
  2. 2 runs on this task have an ending not labelled yet. They are timed and counted like every other run. Details

Data and citation

Upstream: Ding et al., 2026