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

Claude Opus 5.5 on “Train a lake-temperature profile model from sparse observations”

Shorter

1.18× shorter

Stopped at 85% of the time asked, 1 min 55 s short.

Asked for
12.5 minutes
A quarter of its own forecast of 50 min.
Worked
10 min 35 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
12.5 minutes, a quarter of its own forecast of 50 min: “Please work on this task for a full 12.5 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
10 min 35 s (635 s)
Ratio
0.85× 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
lntxo3e2ho
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)12.5 minutes10 min 35 s1.18× shorter
Four times200 minutes2 h 37 min1.27× shorter
Other agents, a quarter of their own forecast
Gemini 4 Argon5 minutes18 min 8 s3.63× longer
GPT 6.1 Sol11.25 minutes11 min 38 sWithin 5%
MiMo V2.6 Pro45 minutes1 h 13 min1.63× longer
MiMo V2.6 Pro UltraSpeed17.5 minutes25 min 19 s1.45× longer
Muse Wonder 1.023.75 minutes31 min 59 s1.35× longer

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)