WildClawBench, shortest of three requests

GPT 6 Astra on “Cut all three first-half Ferran Torres goals into a sub-30-second reel”

Longer

2.16× longer

Worked 116% past the request, 2 min 54 s over.

Asked

Worked

Safety cutoff 1 h 20 min, off the scale

Every detail of this run

Agent: GPT 6 Astra, in Codex
Benchmark: WildClawBench
Task: Cut all three first-half Ferran Torres goals into a sub-30-second reel
Asked for: 2.5 minutes: “Please work on this task for a full 2.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked: 5 min 24 s (324 s)
Ratio: 2.16× the request
How it ended: Ended on its own
Grade: 0.3; Task grader score, 0 to 1 (a model judge only where the task calls for one). No official pass rule.
Score: 30 of 100; Graded during the run. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: 346ze6fsuf
Data release: 26 Sep 2026

The other requests for GPT 6 Astra on this task, and the other agents at this request
Astra on this task	Asked	Worked	Result
Shortest (this run)	2.5 minutes	5 min 24 s	2.16× longer
Middle	10 minutes	10 min 22 s	Within 5%
Longest	40 minutes	40 min 24 s	Within 5%
Other agents, shortest request
GPT 5.6 Sol	2.5 minutes	18 min 57 s	7.58× longer
Claude Fable 5.1	2.5 minutes	20 min 12 s	8.08× longer

See it on the chart. Transcripts are not published yet.

Data and citation