CORE-Bench v1.1, AgentTime v1.1, asked for four times its own forecast

Claude Opus 5.5 on “Human Activities Recognition on PAMAP2 dataset using Random Forest”

Shorter

1.50× shorter

Stopped at 67% of the time asked, 26 min 40 s short.

Asked for
80 minutes
Four times its own forecast of 20 min.
Worked
53 min 20 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
80 minutes, four times its own forecast of 20 min: “Please work on this task for a full 80 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
53 min 20 s (3,200 s)
Ratio
0.67× 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
f2pshbxyki
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
Quarter5 minutes12 min 3 s2.41× longer
Four times (this run)80 minutes53 min 20 s1.50× shorter
Other agents, four times their own forecast
Gemini 4 Argon32 minutes32 min 49 sWithin 5%
GPT 6.1 Sol180 minutes3 h 1 minWithin 5%
MiMo V2.6 Pro240 minutes3 min 49 s62.99× shorter
MiMo V2.6 Pro UltraSpeed280 minutes9 min 47 s28.63× shorter
Muse Wonder 1.0140 minutes31 min 2 s4.51× shorter

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

Data and citation

Upstream: Nadgir et al., 2026