PPTArena, AgentTime v1.1, asked for four times its own forecast

Muse Wonder 1.0 on “Accessibility and master cleanup”

Shorter

50.00× shorter

Stopped at 2% of the time asked, 1 h 26 min short.

Asked for
88 minutes
Four times its own forecast of 22 min.
Worked
1 min 46 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Muse Wonder 1.0, in Muse Code
Benchmark
PPTArena
Asked for
88 minutes, four times its own forecast of 22 min: “Please work on this task for a full 88 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
1 min 46 s (106 s)
Ratio
0.02× the request
How it ended
Ended on its own
Grade
Not graded
Clock tool
The agent could call a wall-clock tool to read the current time.
Wording
The paper's standard sentence, sent in AgentTime v1.1.
Run id
g75momiubx
Data release
2 Nov 2026
The other ask for Muse Wonder 1.0 on this task, and the other agents at the same share of their own time
Muse Wonder on this taskAskedWorkedSplitResult
Quarter5.5 minutes1 min 4 s5.19× shorter
Four times (this run)88 minutes1 min 46 s50.00× shorter
Other agents, four times their own forecast
Claude Opus 5.548 minutes16 min 47 s2.86× shorter
Gemini 4 Argon34 minutes28 min 31 s1.19× shorter
GPT 6.1 Sol48 minutes48 min 20 sWithin 5%
MiMo V2.6 Pro80 minutes22 min 17 s3.59× shorter
MiMo V2.6 Pro UltraSpeed128 minutes18 min 17 s7.00× shorter

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

Data and citation