Terminal-Bench 4.0, AgentTime v1.1, asked for a quarter of its own forecast

Muse Wonder 1.0 on “Rebuild a retired production risk scorer from black-box behavior”

Shorter

2.44× shorter

Stopped at 41% of the time asked, 13 min 17 s short.

Asked for
22.5 minutes
A quarter of its own forecast of 1 h 30 min.
Worked
9 min 13 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
22.5 minutes, a quarter of its own forecast of 1 h 30 min: “Please work on this task for a full 22.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
9 min 13 s (553 s)
Ratio
0.41× 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
b6dwqihxth
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)22.5 minutes9 min 13 s2.44× shorter
Four times360 minutes13 min 14 s27.21× shorter
Other agents, a quarter of their own forecast
Claude Opus 5.513.75 minutes3 min 44 s3.68× shorter
Gemini 4 Argon5 minutes19 min 41 s3.94× longer
GPT 6.1 Sol12.5 minutes12 min 41 sWithin 5%
MiMo V2.6 Pro30 minutes1 h 18 min2.59× longer
MiMo V2.6 Pro UltraSpeed37.5 minutes17 min 44 s2.12× shorter

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)