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

Muse Wonder 1.0 on “Make a Kafka payments worker recover quickly without losing notifications”

Within 5%

Within 5%

Worked 101% of the time asked, within 5% of it.

Asked for
21.25 minutes
A quarter of its own forecast of 1 h 25 min.
Worked
21 min 33 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
21.25 minutes, a quarter of its own forecast of 1 h 25 min: “Please work on this task for a full 21.25 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
21 min 33 s (1,293 s)
Ratio
1.01× 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
7i4psz4myz
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)21.25 minutes21 min 33 sWithin 5%
Four times340 minutes44 min 29 s7.64× shorter
Other agents, a quarter of their own forecast
Claude Opus 5.510 minutes52 min 35 s5.26× longer
Gemini 4 Argon5 minutes45 min 16 s9.05× longer
GPT 6.1 Sol7.5 minutes7 min 50 sWithin 5%
MiMo V2.6 Pro21.25 minutes42 min 46 s2.01× longer
MiMo V2.6 Pro UltraSpeed43.75 minutes17 min 47 s2.46× shorter

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)