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

Muse Wonder 1.0 on “Add scoped state data to state-machine callbacks and history”

Shorter

37.88× shorter

Stopped at 3% of the time asked, 12 h 59 min short.

Asked for
800 minutes
Four times its own forecast of 3 h 20 min.
Worked
21 min 7 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
DeepSWE v1.1
Asked for
800 minutes, four times its own forecast of 3 h 20 min: “Please work on this task for a full 800 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
21 min 7 s (1,267 s)
Ratio
0.03× 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
b7uhvquzss
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
Quarter50 minutes19 min 19 s2.59× shorter
Four times (this run)800 minutes21 min 7 s37.88× shorter
Other agents, four times their own forecast
Claude Opus 5.5200 minutes2 h 14 min1.49× shorter
Gemini 4 Argon152 minutes1 h 6 min2.29× shorter
GPT 6.1 Sol260 minutes4 h 21 minWithin 5%
MiMo V2.6 Pro18 hours40 min 15 s26.83× shorter
MiMo V2.6 Pro UltraSpeed24 hours38 min 25 s37.48× shorter

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

Data and citation

Upstream: Huang et al., 2026