Sakana ALE-Bench, AgentTime v1.1, asked for four times its own forecast

Muse Wonder 1.0 on “AtCoder Heuristic Contest 31”

Shorter

191.01× shorter

Stopped at 1% of the time asked, 11 h 16 min short.

Asked for
680 minutes
Four times its own forecast of 2 h 50 min.
Worked
3 min 34 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Muse Wonder 1.0, in Muse Code
Asked for
680 minutes, four times its own forecast of 2 h 50 min: “Please work on this task for a full 680 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
3 min 34 s (214 s)
Ratio
0.01× 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
cljtmawj5b
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
Quarter42.5 minutes1 min 22 s31.02× shorter
Four times (this run)680 minutes3 min 34 s191.01× shorter
Other agents, four times their own forecast
Claude Opus 5.5360 minutes5 h 2 min1.19× shorter
Gemini 4 Argon180 minutes21 min 31 s8.37× shorter
GPT 6.1 Sol200 minutes3 h 21 minWithin 5%
MiMo V2.6 Pro17.3 hours2 h 1 min8.57× shorter
MiMo V2.6 Pro UltraSpeed20 hours2 h 48 min7.13× shorter

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

Data and citation

Upstream: Imajuku et al., 2025