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

Claude Opus 5.5 on “Generate physically valid walking, jumping, and running trajectories”

Longer

1.20× longer

Worked 20% past the request, 7 min 56 s over.

Asked for
40 minutes
A quarter of its own forecast of 2 h 40 min.
Worked
47 min 56 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Claude Opus 5.5, in Claude Code
Asked for
40 minutes, a quarter of its own forecast of 2 h 40 min: “Please work on this task for a full 40 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
47 min 56 s (2,876 s)
Ratio
1.20× 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
dmobmpw3dv
Data release
2 Nov 2026
The other ask for Claude Opus 5.5 on this task, and the other agents at the same share of their own time
Opus 5.5 on this taskAskedWorkedSplitResult
Quarter (this run)40 minutes47 min 56 s1.20× longer
Four times640 minutes2 h 32 min4.21× shorter
Other agents, a quarter of their own forecast
Gemini 4 Argon12.5 minutes57 min 44 s4.62× longer
GPT 6.1 Sol22.5 minutes22 min 47 sWithin 5%
MiMo V2.6 Pro1.42 hours1 h 16 min1.12× shorter
MiMo V2.6 Pro UltraSpeed1.25 hours28 min 17 s2.65× shorter
Muse Wonder 1.01.25 hours23 min 1 s3.26× shorter

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)