Humanity's Last Exam, AgentTime v1.1, asked for four times its own forecast

Claude Opus 5.5 on “Humanity's Last Exam item 38”

Longer

1.99× longer

Worked 99% past the request, 1 min 35 s over.

Asked for
96 seconds
Four times its own forecast of 24 s.
Worked
3 min 11 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
96 seconds, four times its own forecast of 24 s: “Please work on this task for a full 96 seconds. Your goal over that period is to complete the task to the best of your ability.”
Worked
3 min 11 s (191 s)
Ratio
1.99× 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
74zbcjsv6q
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
Quarter6 seconds2 min 2 s20.28× longer
Four times (this run)96 seconds3 min 11 s1.99× longer
Other agents, four times their own forecast
Gemini 4 Argon96 seconds2 min 29 s1.55× longer
GPT 6.1 Sol120 seconds2 min 14 s1.12× longer
MiMo V2.6 Pro120 seconds1 min 2 s1.95× shorter
MiMo V2.6 Pro UltraSpeed240 seconds4 s57.14× shorter
Muse Wonder 1.0120 seconds1 min 49 s1.10× shorter

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

Data and citation

Upstream: Phan et al., 2025