DeepSWE v1.1, AgentTime v1.1, asked for a quarter of its own forecast

Muse Wonder 1.0 on “tomlkit-toml-table-converters”

Shorter

1.89× shorter

Stopped at 53% of the time asked, 17 min 38 s short.

Asked for
37.5 minutes
A quarter of its own forecast of 2 h 30 min.
Worked
19 min 52 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
37.5 minutes, a quarter of its own forecast of 2 h 30 min: “Please work on this task for a full 37.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
19 min 52 s (1,192 s)
Ratio
0.53× 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
4vrkwaruvv
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
Quarter (this run)37.5 minutes19 min 52 s1.89× shorter
Four times600 minutes18 min 52 s31.80× shorter
Other agents, a quarter of their own forecast
Claude Opus 5.510 minutes49 min 10 s4.92× longer
Gemini 4 Argon4.5 minutes20 min 48 s4.62× longer
GPT 6.1 Sol8 minutes8 min 17 sWithin 5%
MiMo V2.6 Pro30 minutes26 min 46 s1.12× shorter
MiMo V2.6 Pro UltraSpeed33.75 minutes9 min 29 s3.56× shorter

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

Data and citation

Upstream: Huang et al., 2026