TUA-Bench, middle of three requests

GPT 6 Astra on “Fix MP3 metadata”

Longer

1.08× longer

Worked 8% past the request, 24 s over.

Asked

Worked

Safety cutoff 40 min, off the scale

Every detail of this run

Agent: GPT 6 Astra, in Codex
Benchmark: TUA-Bench
Task: Fix MP3 metadata
Asked for: 5 minutes: “Please work on this task for a full 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: 1.08× the request
How it ended: Ending not labelled yet
Grade: Passed; It completed the task in full. The grader returned 1. Pass rule: Full completion: a reward of 1 on the final state.
Score: 100 of 100; Graded during the run. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: 3u2af6ahu5
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	1.25 minutes	1 min 32 s	1.23× longer
Middle (this run)	5 minutes	5 min 24 s	1.08× longer
Longest	20 minutes	20 min 16 s	Within 5%
Other agents, middle request
GPT 5.6 Sol	5 minutes	5 min 45 s	1.15× longer
Claude Fable 5.1	5 minutes	2 min 57 s	1.69× shorter
Muse Spark 1.3	5 minutes	5 min 48 s	1.16× longer

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

Read before quoting. How this run ended has not been labelled yet. It is timed like every other run and counted in every number. Details

Data and citation