TUA-Bench, longest of three requests

Muse Spark 1.3 on “Place heater for sensors”

Shorter

1.92× shorter

Stopped at 52% of the time asked, 28 min 47 s short.

Asked

Worked

Safety cutoff 2 h, off the scale

Asked for: 60 minutes
Worked: 31 min 13 s
Grade: Passed; Score 100 of 100. Graded during the run. Not in the paper's scores.
How it ended: Stopped by an error

Every detail of this run

Agent: Muse Spark 1.3, in Muse Code
Benchmark: TUA-Bench
Task: Place heater for sensors
Asked for: 60 minutes: “Please work on this task for a full 60 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked: 31 min 13 s (1,873 s)
Ratio: 0.52× the request
How it ended: Stopped by an error
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. Not in the paper's scores.
Wording: The standard sentence, used from 4 September 2026.
Run id: u755fbry7o
Data release: 26 Sep 2026

The other requests for Muse Spark 1.3 on this task, and the other agents at this request
Muse on this task	Asked	Worked	Result
Shortest	4 minutes	12 min 18 s	3.08× longer
Middle	15 minutes	no run
Longest (this run)	60 minutes	31 min 13 s	1.92× shorter
Other agents, longest request
GPT 6 Astra	60 minutes	1 h	Within 5%
GPT 5.6 Sol	60 minutes	1 h 1 min	Within 5%
Claude Fable 5.1	60 minutes	26 min 57 s	2.23× shorter

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

Read before quoting. This run was stopped by an error, so its time is a lower bound: the agent might have kept going. Details

Data and citation