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

Muse Wonder 1.0 on “Run the canonical one-year startup simulation with seed 4”

Shorter

41.30× shorter

Stopped at 2% of the time asked, 6 h 11 min short.

Asked for
380 minutes
Four times its own forecast of 1 h 35 min.
Worked
9 min 12 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
YC-Bench
Asked for
380 minutes, four times its own forecast of 1 h 35 min: “Please work on this task for a full 380 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
9 min 12 s (552 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
ps6jpqgpk3
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
Quarter23.75 minutesno run
Four times (this run)380 minutes9 min 12 s41.30× shorter
Other agents, four times their own forecast
Claude Opus 5.5440 minutes2 h 36 min2.82× shorter
Gemini 4 Argon100 minutes2 h 5 min1.25× longer
GPT 6.1 Sol200 minutes3 h 28 minWithin 5%
MiMo V2.6 Pro320 minutes1 h 25 min3.76× shorter
MiMo V2.6 Pro UltraSpeed288 minutes38 min 58 s7.39× shorter

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

Data and citation

Upstream: He et al., 2026